Objective. We sought to quantify the strength of associations between each of four specific developmental disabilities (DDs) and specific types of major birth defects.
Methods. We linked data from 2 independent surveillance systems, the Metropolitan Atlanta Congenital Defects Program and the Metropolitan Atlanta Developmental Disabilities Surveillance Program. Children with major birth defects (n = 9142; born 1981–1991 in metro Atlanta) and 3- to 10-year-old children who were born between 1981 and 1991 in metro Atlanta and identified between 1991 and 1994 as having mental retardation, cerebral palsy, hearing impairment, or vision impairment (n = 3685) were studied. Prevalence ratio (PR), which is the prevalence of a DD in children with 1 or more major birth defects divided by the prevalence of the same DD in children without major birth defects, was measured.
Results. Among the 9142 children who were born with a major birth defect, 657 (7.2%) had a serious DD compared with 0.9% in children with no major birth defect, yielding a PR of 8.3 (95% confidence interval: 7.6–9.0). In general, the more severe the DD, the higher was the PR. Birth defects that originated in the nervous system and chromosomal defects resulted in the highest PRs for a subsequent DD. For all other categories of birth defects, PRs were lowest when all major birth defects present were confined to a single category (ie, isolated defects). PRs for any DD increased monotonically with the number of coded birth defects per child or the number of different birth defect categories per child, regardless of the severity of the defect or whether defects of the nervous system, chromosomal defects, or “other syndromes” were counted.
Conclusions. These data highlight the possible early prenatal origins of some DDs and suggest that both the number of coded birth defects present and the number of anatomic systems involved are strongly related to functional outcomes.
- birth defects
- developmental disabilities
- mental retardation
- cerebral palsy
- hearing impairment
- vision impairment
- surveillance systems
- DD =
- developmental disabilities •
- MR =
- mental retardation •
- CP =
- cerebral palsy •
- TORCH =
- toxoplasmosis-rubella-cytomegalovirus-herpes simplex •
- MACDP =
- Metropolitan Atlanta Congenital Defects Program •
- MADDSP =
- Metropolitan Atlanta Developmental Disabilities Surveillance Program •
- ICD-9 =
- International Classification of Diseases-Ninth Revision •
- HI =
- hearing impairment •
- VI =
- vision impairment •
- PR =
- prevalence ratio •
- CI =
- confidence interval
Developmental disabilities (DDs) comprise a group of nonprogressive conditions first manifested in childhood that result from insults to the developing brain or sensory organs and are associated with deficits in many areas of day-to-day functioning, such as communication, learning, behavior, and motor ability. DDs have been reported to be present in 17% of US children.1 The most serious of the DDs (mental retardation [MR], cerebral palsy [CP], epilepsy, autism, permanent bilateral hearing loss, and legal blindness) together affect 1% to 2% of young children.2–4 These conditions prove costly in terms of special education services, medical and supportive care, demands on caregivers, and economic loss to society.5–9 Furthermore, epilepsy, CP, and MR are associated with a substantially increased risk of premature mortality.10–15
Although the causes for some DDs are well known (eg, Down syndrome, fragile X, toxoplasmosis-rubella-cytomegalovirus-herpes simplex [TORCH] infections, certain structural brain defects),16causes for the majority remain unexplained.17–20 Some structural birth defects, other than those that directly affect the central nervous system, may play a role in the origins of some DDs.21–27 If certain structural birth defects are found to be strongly related to specific DDs, then we might discover additional clues to the prenatal origin of some DDs. These defects also could serve as markers for early identification of children who eventually may need additional medical, educational, and social services. In the present study, the existence of 2 independent, population-based surveillance systems that cover the same geographic area provided a unique opportunity to examine associations between many different types of structural birth defects and each of 4 serious DDs. We are not aware of any previous studies that used a similar methodology.
The data for this report were derived from the Metropolitan Atlanta Congenital Defects Program (MACDP) and the Metropolitan Atlanta Developmental Disabilities Surveillance Program (MADDSP). Both systems cover the 5 Georgia counties of Clayton, Cobb, Dekalb, Fulton, and Gwinnett, an area that includes the city of Atlanta. This area had approximately 33 000 resident live-born infants annually during the study period, 39% of which were black and the remainder almost all white. Both systems have been described in detail elsewhere.2 ,28
The MACDP, which began in 1968, identifies children with major structural birth defects among live or stillborn infants whose mother's residence at the time of birth was within the surveillance area. MACDP personnel find cases from regular chart reviews at the 18 newborn and 2 pediatric referral hospitals that serve the surveillance area, from specialized diagnostic centers, and from vital records.
For this report, children who were born alive to residents of the surveillance area from 1981 through 1991 and included in the MACDP were chosen as the starting point for defining the population of children with birth defects (n = 13 875). Because children must be at least 3 years old to be included in the MADDSP, we excluded children in the MACDP who were known to have died during the first year of life (n = 1024 infant deaths). In addition, we excluded 56 children who had any of 8 selected birth defects that are associated with poor survival rates (trisomy 13, trisomy 18, anencephaly, hypoplastic left heart syndrome, bilateral renal agenesis, alobar holoprosencephaly, thanatophoric dwarfism, and harlequin ichthyosis), 273 children whose birth defects were coded as “probable,” and 3380 children who had exclusively prematurity-related anomalies (eg, patent ductus arteriosus with a birth weight <2500 g) or exclusively minor anomalies (eg, hernias, skin tags) that are no longer considered cases by the MACDP. The final study group included the remaining 9142 children who had at least 1 major birth defect.
In the MACDP, up to 24 different birth defects can be coded for each child. We reviewed all of these codes and classified each of the 9142 study children into 1 or more of 11 anatomic or etiologicInternational Classification of Diseases–Ninth Revision (ICD-9) categories on the basis of their major birth defects (not considering any coexisting minor or “probable” codes), with many children being included in more than 1 category. The “nervous system” category was subdivided further into neural tube defects (eg, spina bifida, encephalocele) and all other nervous system defects. The “eye, ear, other head and neck” category was subdivided into 3 subcategories: “eye,” “ear,” and “other head and neck” defects. The chromosomal defects category was subdivided into Down syndrome and all other chromosomal defects. The “all other major defects” category included “other syndromes and specified anomalies” (ICD-9 759.6, 759.8, 760.71, 760.75); congenital TORCH infections; neoplasms; endocrine, nutritional, metabolic, and immune system defects; and all remaining defects (primarily those coded in the ICD-9 759 series that were not included in another category). Some single-gene disorders may be included in this category. We identified children whose major birth defects were confined to only 1 of the 11 birth defect categories, for which we used the term “isolated defects.” Sequences were not identified easily as such and therefore may or may not be included as isolated defects.
The MADDSP, begun in 1991, identifies children who are aged 3 to 10 years and have MR, CP, serious bilateral hearing impairment (HI), or serious bilateral vision impairment (VI) and whose parents are residents of the surveillance area when the children are identified. Surveillance case definitions for these 4 conditions have been published.2 Children who have these conditions are found through review of records at multiple public and private sources, with the special education departments in the 9 public school systems that serve the surveillance area being the major source. For each surveillance year, we compared electronically the list of children enrolled in the MADDSP with the MACDP file to identify matches. Questionable matches are checked manually, and the MACDP data are appended to the MADDSP file for all matched children.
DDs in children who were enrolled in the MADDSP during 1991 to 1994 and born to a resident of the study area from 1981 to 1991 served as the outcome variables (n = 3685 children). Of this total, 3031 children had MR, 796 had CP, 286 had HI, and 270 had VI; 558 (15%) children had 2 or more of the 4 DDs. We subdivided children with MR or CP according to severity and the presence of coexisting DDs. For MR, we used the commonly defined categories of mild MR (IQ 50–70) and severe MR (IQ <50). We categorized children with CP as “disabled” or “nondisabled” on the basis of published criteria.29We defined children as having “isolated MR” or “isolated CP” if they did not have a coexisting disability, including autism or a seizure disorder. Although autism and seizure disorders are not among the conditions surveyed by the MADDSP, they are included in the database if found during record review as coexisting with any of the 4 target conditions.
The prevalence of DDs in children with a major birth defect (n = 9142) was compared with the prevalence of DDs among all 1-year survivors who were born to residents of the study area during 1981 to 1991, excluding those with a major birth defect (n = 358 051 − 9142 = 348 909). The ratio of these rates (prevalence ratio [PR]) was calculated as the measure of the strength of the association between major birth defects and DDs. We subdivided children with major birth defects both by the number of birth defect codes that they had and by the number of anatomic ICD-9 categories encompassed by their birth defects. We computed 95% confidence intervals (CIs) for the PRs and linear trend statistics (Mantel test) using the Statistical Analysis Battery for Epidemiologic Research.
Of the 9142 study children who were born with 1 or more major birth defects, 657 (7.2%) were identified independently by the MADDSP as having a DD at ages 3 to 10 years while they were still living in the surveillance area. Conversely, 657 (17.8%) of the 3685 children who were aged 3 to 10 years and had a DD had been identified independently with a major birth defect by the MACDP. Of the 657 children with both a major birth defect and a DD, 390 (59%) had either defects of the nervous system or chromosomal defects alone, or in combination with other defects.
Compared with the prevalence of DDs among the 348 909 children who had no major birth defects and who survived the first year of life, the prevalence of DDs among the 9142 children with major birth defects was exceptionally high (Table 1). The PRs varied from 4.5 for isolated CP to 21.8 for isolated severe MR. The PR for severe MR was higher than for mild MR, and the presence of another DD did not necessarily increase the strength of the associations with MR. However, the PR was much higher for disabling CP than for nondisabling CP and for CP coexisting with another DD than for isolated CP. VI was more strongly associated with major birth defects than was HI.
The strength of the association between birth defects and DDs varied with type of birth defect and presence of other types of birth defects (Table 2). The strongest associations were evident for chromosomal defects (PR = 62.5); defects of the nervous system (PR = 30.2); and eye, ear, and head and neck defects (PR = 20.4), and the weakest association prevailed for genital organ defects (PR = 4.8). Although the magnitude of the associations with DD generally were weaker when children with either nervous system or chromosomal defects were eliminated, they still were substantial. With the exception of nervous system defects and chromosomal defects, the associations between DDs and specific birth defect categories were weakest when the analyses were limited to children with isolated defects. In those analyses, the prevalence of DDs was still statistically elevated in all birth defect categories, except for integument, for which there were no DD cases.
Each of the 4 DDs was strongly associated with most specific categories of isolated birth defects (Table 3). The most consistent and strongest associations were between each of the 4 DDs and “other nervous system” defects, which were related primarily to structural brain anomalies. In several instances, associations were unusually strong between a specific DD and a specific isolated birth defect, for example, MR and eye defects (PR = 7.2), CP and cardiovascular defects (PR = 7.3), HI and urinary tract defects (PR = 16.5), and VI and neural tube defects (PR = 43.0). Within the “all other major defects” category (data not shown), we found the expected associations between MR and congenital TORCH infections (PR = 24.3)30 and between MR and fetal alcohol syndrome (PR = 29.1).31 The high prevalence of MR (PR = 23.3) in children with defects classified as “other syndromes” (ICD-9, 759.6, 759.8) confirms previous clinical case reports of such associations.32
The PR for any DD increased monotonically with the number of coded birth defects per child (Fig 1) and with the number of birth defect categories represented by a child's major defects (Table 4). The rising trend in PRs with the number of birth defects or the number of birth defect categories was attenuated but remained strong when children with nervous system defects, chromosomal defects, or “other syndromes” were excluded from these analyses.
This report suggests a role for deleterious factors operating prenatally in the origins of many DDs. In some children, 1 or more birth defects may be the underlying cause of a DD (such as Down syndrome), and in other children both their birth defect(s) and their DD may be because of a common third factor (eg, prenatal alcohol exposure). Although children with a major birth defect may undergo more intense scrutiny than other children later in life by medical and educational professionals and thus be more likely to have a DD diagnosed, the range and magnitude of increased risks seem to dispel this phenomenon as an explanation for the findings. Clearly, the magnitude of the increased risk associated with major birth defects is not uniform over specific subtypes of DDs. PRs were lowest for mild MR and for isolated or nondisabling CP and highest for isolated severe MR and for VI. Future, more detailed analyses could focus on the specific DDs with the strongest associations. In terms of attributable fractions, major birth defects may be associated with as many as 16% of all children with 1 or more of the 4 DDs studied here and with approximately 35% of all cases of isolated severe MR.
Our data indicate that the absolute number of recorded birth defects codes and the number of anatomic systems involved are predictive of functional outcomes. These findings suggest that the number of codes may be a marker of the severity of the defect and the pervasiveness of the damage to the developing fetus (even if only 1 defect may be present). It also is possible that some children with many coded birth defects are those with unrecognized syndromes (which are underascertained in the MACDP) that are strongly related to some DDs.
Some of the birth defects examined here probably caused 1 or more DDs directly by disrupting normal functional development. For example, nervous system defects caused some MR or CP, ear defects caused some HI, and eye defects caused some VI. The specific associations between these birth defects and DDs are exceptionally strong (PR: 20.0–133.3). However, each of the 4 DDs also was strongly associated with isolated birth defects that would not be expected to alter functional development of those organs affected by the DDs. For example, MR was strongly associated with isolated eye defects as well as isolated defects of the cardiovascular, respiratory, gastrointestinal, reproductive, urinary, and musculoskeletal systems. These multiple associations suggest that some cases of MR are not caused directly by coexisting birth defects but instead may be produced by other factors present early in embryonic development, which act as common causes of both birth defects and MR. Alcohol and isotretinoin are examples of such teratogens.31 ,33 Alternatively, the timing of an insult during pregnancy may determine whether a birth defect and a neurobehavioral disorder are produced in the same child. For example, alcohol consumption starting early in pregnancy can induce dysmorphologic features, whereas the same exposure starting late in pregnancy results in mental and behavioral deficits.34 ,35
Some of the DDs included here could have been caused by postnatal events, such as strokes, injuries, and infectious diseases. In fact, for the 1991 MADDSP surveillance year, an estimated 4.5% of all DDs resulted from postnatal events.36 Other DDs may have been caused by perinatal events, such as infections, asphyxia, and preterm delivery.16 ,17 If we excluded from our data set DDs known to be postnatal or perinatal in origin, then the association between birth defects and the remaining DDs would be even stronger.
The PRs shown here can be viewed as estimates of what the “true” PRs would have been if all children, with or without major birth defects, in the birth cohorts under study had been followed completely from age 1 year to 10 years to identify all cases of the selected DDs. In the present study, however, not all children in the birth cohorts of interest had attained their 10th birthday during the 1991 to 1994 surveillance years, others may have died with a DD before being eligible to be ascertained by MADDSP, and for others, 1 of the selected DDs may have been diagnosed only after their families emigrated from the study area. Because we did not have information on deaths that occurred among cohort members who survived infancy, we attempted to account for some of the deaths by excluding from the analysis defects known to have poor survival rates. Selective emigration (ie, children with birth defects being more or less likely than children without birth defects to leave Atlanta before the recognition of their DD) may have occurred. Nevertheless, given the strength of the associations that we found between the specific birth defects and the DDs, it seems unlikely that these methodologic shortcomings could explain fully the associations that we reported.
This report highlights the role of prenatal factors in the origin of many DDs and suggests that a sizable proportion of DDs may be caused by insults that occur early in embryologic development. Because many cases of DDs are not manifested until months or years after birth, both the type and the number of major birth defects present in the first year of life may serve as a marker or early warning sign of subsequent cognitive, motor, and sensory deficits. Our data suggest that birth defects pose a greater burden on society than previously recognized.
We thank Dr Cynthia Moore for insightful comments, which helped to strengthen this article.
- Received August 4, 2000.
- Accepted March 14, 2001.
Reprint requests to (C.A.B.) Centers for Disease Control and Prevention (F-15), 4770 Buford Hwy NE, Atlanta, GA 30341. E-mail:
- Boyle CA,
- Decouflé P,
- Yeargin-Allsopp M
- ↵Boyle CA, Yeargin-Allsopp M, Doernberg NS, Holmgreen P, Murphy CC, Schendel DE. Prevalence of selected developmental disabilities in children 3–10 years of age: the Metropolitan Atlanta Developmental Disabilities Surveillance Program, 1991. MMWR Morb Mortal Wkly Rep. 1996;45(SS-2):1–14
- ↵Yeargin-Allsopp M, Murphy CC, Oakley GP, Sikes K. A multiple-source method for studying the prevalence of developmental disabilities in children: the Metropolitan Atlanta Developmental Disabilities Study.Pediatrics. 1992;89:624–630 (published erratum appears inPediatrics. 1992;90:1001)
- ↵Birenbaum A, Guyot D, Cohen HJ. Health Care Financing for Severe Developmental Disabilities. Washington, DC: American Association on Mental Retardation; 1990
- ↵Braddock D, Hemp R, Parish S, Westrich J. The State of the States in Developmental Disabilities. 5th ed. Washington, DC: American Association on Mental Retardation; 1998
- ↵National Foundation for Brain Research. The Cost of Disorders of the Brain. Washington, DC: Lewin-ICF; 1992
- ↵Waitzman NJ, Scheffler RM, Romano PS. The Costs of Birth Defects: Estimates of the Value of Prevention. Lanham, MD: University Press of America; 1996
- ↵US Department of Education. Twentieth Annual Report to Congress on the Implementation of the Individuals with Disabilities Education Act. Washington, DC: US Department of Education; 1998
- Evans PM,
- Evans SJW,
- Alberman E
- Hutton JL,
- Cooke T,
- Pharaoh POD
- ↵Dupont A, Mortensen PB. Avoidable death in a cohort of severely mentally retarded. In: Fraser WI, ed. Key Issues in Mental Retardation Research. London, England: Routledge; 1990:28–34
- Anderson G,
- Schroer RJ,
- Stevenson RE
- Pinto-Martin JA,
- Riolo S,
- Cnaan A,
- Holzman C,
- Susser M,
- Paneth N
- ↵Jones KL. Smith's Recognizable Patterns of Human Malformation. 5th ed. Philadelphia, PA: WB Saunders; 1997
- ↵Adams J, Lammer EJ. Relationship between dysmorphology and neuro-psychological function in children exposed to isotretinoin “in utero.” In: Fujii T, Boer GJ, eds. Functional Neuroteratology of Short-Term Exposure to Drugs. Tokyo, Japan: Tokyo University Press; 1991:159–168
- Becker HC,
- Randall CL,
- Salo AL,
- Saulnier JL,
- Weathersby RT
- Coles C
- Copyright © 2001 American Academy of Pediatrics