Introduction to Developmental and Behavioral Screening

Description

Introduction to screening young children for development and behavior problems in primary care, emphasizing standardized tools.

Learning objectives

Impact of screening

Understand the potential of developmental and behavioral screening in primary care settings in reducing the severity and impact of these problems on children and families.

Know specific screening instruments

Be aware of specific screening instruments with appropriate sensitivity, specificity, and ease of use in primary care.

Screening versus referral

Identify who should be screened and who should be referred in a primary care setting.

Under-referral in screening

Identify pitfalls of screening that may lead to under-referral.

Early detection of children with developmental, behavioral, and emotional delays, risks and disabilites problems is an important responsibility for pediatricians. This module provides information about techniques that can be used effectively and efficiently in the office setting. We provide background information on screening and surveillance, and an annotated list of quality tools.

Using this learning module

Read this introduction to Developmental and Behavioral detection, then look at the the look at related items in the Screening section of dbpeds.org. You will find many resources and practical articles on methods for early detection as well as links to tools mentioned in the article. We also have many parent handouts available related to screening.

Overview

About 16% to 18% of children have disabilities such as speech-language impairments, mental retardation, learning disabilities, and emotional/behavioral disturbance. Although such children are twice as likely to seek health care as children without disabilities, only 20% to 30% of these children are detected prior to school entrance. Under-enrollment rates in early intervention programs (of roughly 80%) confirm the need to improve early detection in primary care. Still, the reasons are understandable: Many disabilities are subtle and children who have them may appear to be developing normally especially at younger ages. Few providers are familiar with measures workable in busy clinics. Reimbursement has not consistenly kept pace with the cost of services.

Under-detection is nevertheless unfortunate because it may delay the possibilities of early intervention. Children who participate in early intervention programs prior to kindergarten are more likely to graduate from high school, hold jobs, live independently, avoid teen pregnancy, delinquency and violent crime. These positive outcomes save society between $30,000 and $100,000 per child. Put another way, for every $1.00 we spend on early intervention, society saves $13.00 - a savings so substantial that countries such as Great Britain fund early intervention programs directly from their national treasury and thus offer quality of early preschool program to all low-income children as well as to those with delays and disabilities.

Policy Statements

The American Academy of Pediatrics recognizes the benefits of early detection and thus earlier intervention. In its July, 2006 policy statement (Pediatrics, 2006), members of the Council on Children with Disabilities wrote:

"Early identification of developmental disorders is critical to the well-being of children and their families. It is an integral function of the primary care medical home and an appropriate responsibility of all pediatric health care professionals."

The American Academy of Neurology, in its Practice Parameter on diagnosis of autism (2000), stated:

"[Early detection] should be performed at all well-child visits from infancy through school-age, and at any age thereafter if concerns are raised about social acceptance, learning, or behavior."

Both the AAP and the AAN exhort providers to engage in developmental surveillance. Surveillance is a longitudinal process that commences with routinely eliciting and addressing parents' concerns, followed by reviewing medical history, maintaining a record of developmental progress, making accurate and informed observations about the child and parent-child interactions, identifying risk and protective factors that often predict developmental risks or resilience, and ensuring that needed interventions are promptly delivered.

Periodic use of quality screening tools is recommended to add evidence to the process of surveillance. The AAN specifically recommends tools with abundant psychometric support, i.e., national standardization, reliability (of several types), validation against criterion diagnostic tests, and accuracy, i.e., sensitivity and specificity of 70% or greater. Recommended tools include the Ages and Stages Questionnaire, Parents' Evaluation of Developmental Status, and the Brigance Screens. Tools not meeting standards for test construction were specifically excluded (e.g., Denver-II, R-PDQ). Like the AAN, the AAP also supports use of quality tools but provides an extensive table of measures that includes tools with lesser levels of psychometric support. Nevertheless, the AAP statement offers some guidance in how to select quality measures from the extensive array.

Both societies suggest a periodicity schedule for surveillance and screening. Both agree that surveillance should occur at all well-visits from infancy through school age. The AAN also recommends use of screening tools at each visit, while the AAP emphasizes use of screens at at 9, 18, 24 or 30 months but is less clear on whether qualilty screens should be used at subsequent visits. Given that developmental problems can develop past 24 - 30 months of age (e.g., language impairment, learning disabilities) and that the weight of evidence currently rests with quallity screens (e.g, as documented by the Commonwealth Fund's ABCD Initiative in which primary care providers were trained in the use of quality screens, required to use them, reimbursed accordingly, and taught about available services, ensured that the Center for Disease Control's prevalence estimates became commensurate with enrollment rates in early intervention) it is wise that the AAP is conducting implementation research on the effectiveness of its current policy.

In the tutorial on developmental-behavioral screening on www.dbpeds.org, we provide articles on meant to help child health providers improve their efforts in early detection. We focus on current mandates as well as policy and ultimately on measures with proven effectiveness.

EPSDT and MEDICAID

The Early Intervention Periodic Screening, Diagnosis and Treatment benefit under Medicaid, also calls for screening at each well visit. There has been controversy about how this requirement should be implemented. For example, there are currently 28 States engaged in class action law suits against Medicaid in several cases for failing to deliver on the promise of early screening (for a summary, see the EPSDT Fact Sheet).

Three States have settled their suits by agreeing to ensure that providers administer standardized, validated screens: Tennessee, North Carolina, and Louisiana. Others, such as North Carolina, Georgia and Arizona, enjoy collaborative efforts between Medicaid and the State AAP Chapters. These are dramatically improving deployment of quality tools at well-visits and ensuring that providers are reimbursed. Medicaid's new policy allowing providers to undbundle screening from the well-visit through the use of the 96110 procedure code, carry national reimbursement rates averaging $13.00.

Who should be screened?

Screening is designed for the asymptomatic -- for those thought to be developing normally. Children with obvious problems can simply be referred promptly for early intervention. There is broad agreement that there should be a systematic way of identifying children who need to be evaluated for disabilty. There is good evidence to support the use of standardized, validated tools to do this screening. If only symptomatic children are screened, children who might have been identified with screening may be missed.

Screening tools

We include here a description of qualilty screening tools. All improve existing detection rates by 3 to 4 times and will detect correctly almost all children with and without problems when used longitudinally. Tests included in the discussion are those with national standardization, proven reliability, validation against a range of quality diagnostic measures, sensitivity in the detection of children with problems, and specificity in the detection of children without problems in keeping with standards for screening tests of at least 70% to 80%.

For an overview of quality screening tools, including where to get them and how much they cost, see the article Commonly Used Screening Tools on dbpeds.org.

Issues in Screening via Information from Parents

Most parents can complete screens. However, many parents with limited education do not raise concerns spontaneously with health care providers. Such parents may not realize that developmental topics are a part of primary care. Thus, parent-report screens alert families that development is an important topic to providers and ensures that all parents have a voice in a pediatric encounter.

Even so, parents with limited education may have difficulty with literacy. Asking parents, in a sensitive manner, if they would prefer to have someone go through the measure with them should circumvent literacy barriers. Most screens are also published in languages such as Spanish, Vietnamese, French, etc. to help circumvent language barriers.

Some clinicians worry that highly educated parents are likely to be overly concerned about their children and that parent-report tools used with this group will result in over-referrals. Tests correct for this problem in various ways (e.g. by identifying those concerns not predictive of developmental or behavioral problems) so that clinicians can offer patient education instead of referrals. Even so, screens always produce some errors.

However, is important to know that over-referrals on screening tests (false-positives) tend to be on children who perform below average in the better predictors of school success, i.e., intelligence, academics/preacademics, and language skills, and to have numerous psychosocial risk factors. Identifying these children, even though they will not qualify for special education, can help providers link them with needed services such as Head Start, Early Head Start, after-school tutoring, summer school, quality day care and preschool, Boys and Girls’ Clubs, parent training, etc.

Pitfalls of Screening

Here are five major pitfalls of screening:

Waiting until a problem is observable. Clinicians often use screening tools only after they've noticed a problem. If the problem is obvious, referring is the correct response. Screening such children may compound rather than reduce the error in both clinical judgment and screening tools because screening instruments are not intended for children with recognized problems.
Ignoring screening results. Many times, screening test results are not taken seriously and children who fail are not referred but rather viewed with a wait and see attitude. Good screens make a correct decision at least 70% to 80% of the time. Children over-referred by such screens tend to have numerous psychosocial risk factors and perform well below average on diagnostic measures. Thus over-referrals tend to be of children for whom extra attention from nonmedical providers is invaluable. Children under-detected by screens at one visit are likely to be identified at a subsequent visit. Reliability studies conducted as a part of constructing quality screening measures ensure that the results will be virtually identical on a second administration. Thus, there is no evidence to support "temporizing" in such a situation, and harm could result from delayed diagnosis and intervention.
Relying on informal methods. Informal tools such as checklists often contain items sure to miss most kids with problems (such as the checklists on many encounter forms). These are often based on instruments such as the original Denver Developmental Screening Test, that missed up to 50 per cent of mental retardation and 70 per cent of language impairment.

Checklists provide no validated criteria for referral, and have unknown reliability. Remember, most developmental and behavioral problems are not obvious. Children usually walk although some do not walk well. They talk but may not talk well; read but not well. Discriminating between adequately developed and problematic levels of skills requires careful measurement. Validated and standardized tools carry the burden of proof that informal measures lack. We would never select tools for blood lead or other medical screens with questionable or unknown levels of accuracy. Why do we do this with development?
Using a measure not suitable for primary care. Most practices retain copies of tests such as the Denver-II or Gessell. Concerns about the accuracy of both measures aside, these tests also take longer to administer than the average length of a well visit. This results in two things: 1) using tests only with selected patients, usually those with observable problems; and 2) nonstandard administrations, such as administration of key items on the Denver-II. The former situation ignores the asymptomatic who are those most in need of screening while the latter reduces screens to the status of checklists with their inherent lack of criteria and unproven sensitivity and specificity.
Assuming services are limited or nonexistent. Although the early childhood community has not always reached out to medical providers, services of good quality exist in every state for any child with developmental delays. It is a good idea to meet the local coordinators responsible for implementing child-find and intervention programs under the Individuals with Disabilities with Education Act (IDEA). Encourage them to keep you posted on referrals and consider taking time to visit some programs.

For information on programs, the National Early Childhood Technical Assistance Center has a website with contact information for every state and usually region for children 0-3 or 3-5 years. For children 5 years and older, contact a school psychologist in the child's school of zone or your local school board (see Public school locator). Generally, IDEA programs are a one stop shop and can provide not only broad early intervention and special education programs but also many related services including mental health, physical, occupational, and speech therapies, parent training, etc.

It is also important to monitor the development of children who do not qualify for IDEA programs but who still have developmental needs. Most of these children can be well served by a quality day care program (see the National Association for the Education of Young Children website for a list of accredited programs). When parents have limited income, children may be eligible for services through Head Start, Early Head Start or Migrant Worker Head Start.

Summary

There is a broad consensus that early identification of disability is both feasible and important. Improving the effectiveness of developmental screening and surveillance has the potential for improving the lives of countless children and families. There is also good evidence that more effective identification will result in cost savings to society.

Articles in this guide provide examples of successful programs, help to make changes in your practice, and practical advice. Many of the articles are also suitable for trainees and office staff.

Please take a few minutes to give us feedback about the articles as you go through them.

Disclosure: Frances P. Glascoe, co-author of this article, is also author of several screening tools including Parents' Evaluations of Developmental Status (PEDS), PEDS:Developmental Milestones (PEDS:DM), and the Safety Word Inventory and Literacy Screening (SWILS).

References

Glascoe FP: Are over-referrals on developmental screening tests really a problem? Arch Pediatr Adolesc Med 2001; 155:54 http://archpedi.ama-assn.org/issues/v155n1/rfull/poa00202.html [269]
Bethell C, Peck C, Schor E: Bethell C, Peck C, Schor E: Assessing health system provision of well-child care: The Promoting Healthy Development survey. Pediatrics 2001;107:1084 http://www.pediatrics.org/cgi/content/full/107/5/1084 Pediatrics 2001;107:1084 [262]
Pelletier H: Assuring Better Child Health and Development (ABCD) Project, 2002. http://www.nashp.org/ [264]
Halfon N, McLearn KT, Schuster MA (eds): Child Rearing in America: Challenges Facing Parents with Young Children. New York, Cambridge University Press, 2002 http://books.cambridge.org/0521012643.htm [261]
Rappo PD: Coding for mental health and behavioral problems: The arcane elevated to the ranks of the scientific. Pediatrics 2002;110;167 [266]
American Academy of Pediatrics (2001). Developmental surveillance and screening of infants and young children. Pediatrics 108(1): 192-6. [1]
Dobrez D, Lo Sasso A, Holl J, et al: Estimating the cost of developmental and behavioral screening of preschool children in general pediatric practice. Pediatrics 2001;108:913 [268]
Shonkoff, J. P., D. Phillips, et al. (2000). From neurons to neighborhoods : the science of early child development. Washington, D.C., National Academy Press. [189]
International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). Los Angeles, Calif., Practice Management Information Corporation, 2003 [267]
Reynolds AJ, Temple JA, Robertson DL, et al: Long-term effects of an early childhood intervention on educational achievement and juvenile arrest: A 15-year follow-up of low-income children in public schools. JAMA 2001;285:2339 http://jama.ama-assn.org/issues/v285n18/abs/joc01444.html [259]
Gomby DS, Larner MB, Stevenson CS, et al: Long-term outcomes of early childhood programs: Analysis and recommendations. Futures of Children 1995;5:6 http://www.futureofchildren.org/homepage2824/archive.htm [260]
Rushton, J., D. Bruckman, et al. (2002). Primary care referral of children with psychosocial problems. Arch Pediatr Adolesc Med 156(6): 592-8. [185]
Halfon N, Olson L, Inkelas M, et al: Summary statistics from the National Survey of Early Childhood Health, 2000. National Center for Health Statistics. Vital Health Stat 15(4), in press. http://www.cdc.gov/nchs/about/major/slaits/nsech.htm [263]
Glascoe FP, Byrne KE, Chang B, et al: The accuracy of the Denver-II in developmental screening. Pediatrics 1992; 89:1221 http://www.pediatrics.org/ [265]
Wolraich ML, Felice ME, Drotar D (Ed). The classification of child and adolescent mental diagnoses in primary care. [90]
American Academy of Pediatrics (2001). American Academy of Pediatrics. The new morbidity revisited: a renewed commitment to the psychosocial aspects of pediatric care. Committee on Psychosocial Aspects of Child and Family Health. Pediatrics 108(5): 1227-30. [164]
Johnson C: Using Developmental and Behavioral Screening Tests. Pediatr Rev 2000;21:255 http://pedsinreview.aapjournals.org/content/vol21/issue8/index.shtml#ARTICLES [270]