 |
|
Health Care Use by Children Diagnosed as Having Developmental Delay
Margaret M. Gallaher, MD, MPH;
Dimitri A. Christakis, MD, MPH;
Frederick A. Connell, MD, MPH
Arch Pediatr Adolesc Med. 2002;156:246-251.
ABSTRACT
| |
Background Although children with developmental delay are known to have increased
health care use, it is unclear what proportion of that health care use is
related to associated chronic health conditions.
Objectives To assess the prevalence of isolated developmental delay and to determine
the role of developmental delay in health care use controlling for chronic
health conditions.
Design Retrospective cohort study using Washington State Medicaid claims records
from November 1, 1990, to December 31, 1997, an administrative data set that
contains both International Classification of Diseases,
Ninth Revision, Clinical Modification codes and billed services.
Patients and Setting Children born between November 1, 1990, and December 31, 1992, diagnosed
as having developmental delay before the age of 5 years, enrolled in Medicaid
within 1 month of birth, and continuously enrolled for at least 12 months.
Four control subjects per case were matched on date of birth and duration
of continuous enrollment in Medicaid.
Main Outcome Measures Visits to physicians, emergency departments, other practitioners, or
hospitals by year of life.
Results One thousand two hundred forty-two children having developmental delay
and 5370 children without developmental delay were included. One percent of
those who met study criteria had developmental delay without chronic health
conditions and 30% of the children with developmental delay had no associated
chronic health conditions. Boys were 1.6 times as likely to have a diagnosis
of developmental delay. Developmental delay was independently associated with
increased health care use by all 4 measures used.
Conclusion Developmental delay increases health care use apart from associated
chronic health conditions.
INTRODUCTION
ESTIMATED TO constitute between 2% and 10% of the pediatric population,
children with developmental delay have been shown to use more health care
than children without developmental delay.1-5
However, developmental delay is associated with chronic health conditions
that also increase health care use.6 The prevalence
of isolated developmental delay and the role it plays in the use of health
services, apart from comorbid chronic health conditions, is unknown. This
study was designed to determine the prevalence of isolated developmental delay
and to describe the role that developmental delay plays in health care use
in the first 5 years of life controlling for chronic health conditions.
PATIENTS AND METHODS
PATIENTS
We performed a retrospective longitudinal cohort study using Washington
State Medicaid claims data from November 1, 1990, to December 31, 1997, that
contain both encounter data and International Classification
of Diseases, Ninth Revision, Clinical Modification coding. Eligible
children were born between November 1, 1990, and December 31, 1992, and were
diagnosed as having developmental delay (International Classification
of Diseases, Ninth Revision, Clinical Modification codes 315, 317,
318, and 319). To ensure complete information on health care use, we included
only those children who were enrolled in Medicaid within 1 month of birth
and who remained continuously enrolled for at least 12 months thereafter,
as determined by review of the Medicaid eligibility file. The diagnosis of
developmental delay had to occur prior to the first break in Medicaid enrollment
and before the child was 5 years old. Claims submitted before the first lapse
in enrollment, or until the child was 5 years old, whichever came first, were
included in this analysis.
We maximized study power by randomly selecting 4 control subjects per
case, matched within 1 month of the date of birth and by duration of continuous
enrollment in Medicaid. Children selected as controls could not have had a
diagnosis of developmental delay in the Medicaid claims data prior to their
fifth birthday, regardless of any lapse in enrollment.
CHRONIC HEALTH CONDITIONS
To assess the presence of chronic health conditions, International Classification of Diseases, Ninth Revision, Clinical Modification diagnoses were grouped by body system. Included conditions were those
that would be expected to affect a child's physical well-being for 6 months
or longer and necessitate medical and/or surgical intervention. A list of
conditions is available from us on request. We included all diagnoses that
occurred in the claims data during the first period of continuous enrollment,
or until the child was 5 years old, whichever came first. Children who had
one or more diagnoses within a single body system were considered to have
1 chronic health condition. To maximize sensitivity and to ensure a representative
sample of children without chronic health conditions, a diagnosis need only
be present once to classify a child as having that condition.
HEALTH CARE USE
Health care use was the outcome of interest and was assessed based on
the type of health care provider submitting the claim to Medicaid. Health
care providers were grouped as follows: physician (includes outpatient clinic),
hospital, emergency department (ED), and other practitioner (includes physical
therapist, occupational therapist, speech therapist, audiologist, and services
provided by schools). We assessed the number of visits to a physician, other
practitioner, ED, and hospitalizations based on claims submitted to Medicaid.
If 2 different health care providers submitted a claim for the same child
on the same day, 2 visits were counted. Claims for visits by a physician or
other practitioner that occurred during the course of a hospitalization were
counted as part of the hospitalization and, thus, were excluded from the calculations
of use for physician or other practitioner.
We adjusted utilization measures for the total person-time contributed
by each child, since all children were not enrolled for the full 5 years,
including the total months of continuous enrollment from birth to the first
break in enrollment or until the child was 5 years old, whichever was less,
in the denominator. Visits to a physician or outpatient clinic, ED, and other
practitioners are reported as the mean number per child by year of life. Hospitalizations
are reported as the mean number per 100 children by year of life. Because
of incomplete data, hospitalizations in the first year of life include only
those that occurred after the birth hospitalization. In Washington state,
an infant's hospital care at delivery for an uncomplicated birth is covered
by the mother's Medicaid number, so this hospitalization is not recorded on
the infant's Medicaid claims.
ANALYSIS
Analyses of use were stratified by 4 groups: (1) children who had developmental
delay and no chronic health conditions; (2) children who had no developmental
delay and no chronic health conditions; (3) children who had developmental
delay with 1 or more chronic health conditions; and (4) children who had no
developmental delay, but who had 1 or more chronic health conditions. Given
the skewed distribution of visits per child, we compared the mean values using
the nonparametric Kruskal-Wallis test.7 A prevalence
odds ratio (OR) was used to compare categorical variables; the Mantel-Haenszel
OR was used to provide a summary prevalence OR for chronic conditions stratified
by sex.8 We used the Cornfield approximation
for the 95% confidence interval (CI).9
We developed a series of regression models to assess the independent
association of developmental delay on our utilization measures. Because of
the skewness of the distribution of these measures, we used negative binomial
regression to calculate adjusted incident rate ratios (IRRs).10
The dependent variable was the number of visits by each type of health care
provider during the child's first 5 years of life. The independent variable
of interest was developmental delay (0, none; 1, present). Sex (0, female;
1, male), race (white race as reference), and the type of chronic health condition
(0, not present; 1, present) were included in the model to adjust for confounding.
RESULTS
A total of 126 671 children with a birth date between November
1, 1990, and December 31, 1992, were enrolled in Washington State Medicaid.
Of these, 38 244 (30.2%) were enrolled within 1 month of birth and continuously
for at least 12 months. Of these, 1242 (3.2%) were identified as having developmental
delay. A total of 5370 children were randomly selected as controls.
The demographic characteristics of the study population are shown in Table 1. Children having a diagnosis of
developmental delay were more likely to be male, less likely to have race
recorded as Hispanic or other, and more likely to have race recorded as unknown.
Eighty-one percent of the study children were enrolled continuously for 4
to 5 years (Table 1). The mean
(SD) age of the first diagnosis of developmental delay was 33.7 (12.6) months,
with a median of 33 months.
|
|
|
|
Table 1. Demographic Characteristics of Study Population
|
|
|
CHRONIC HEALTH CONDITIONS
Table 2 summarizes the frequency
of chronic health conditions among children with and without developmental
delay. Children with developmental delay were twice as likely to have 1 or
more chronic health conditions. Of the children with developmental delay,
391 (32%) had isolated developmental delay. This is 1% of the 38 244
children who were enrolled in Medicaid at birth and continuously for at least
12 months.
|
|
|
|
Table 2. Number of Chronic Health Conditions Among Children Who Have
and Who Do Not Have a Diagnosis of Developmental Delay
|
|
|
One third of the children with developmental delay had only 1 chronic
health condition in addition to developmental delay (Table 2). Thirty-seven percent had 2 or more chronic health conditions
in addition to developmental delay. Among the children who did not have a
diagnosis of developmental delay, there were twice as many children with no
chronic health conditions compared with those with developmental delay, and
only 9% had 2 or more chronic health conditions (Table 2).
Children with developmental delay were more likely to have every chronic
health condition that we examined. Compared with children without developmental
delay and controlling for sex, children with developmental delay were more
than 9 times as likely to have a centeral nervous system condition; nearly
9 times as likely to have an ear, nose, and throat condition; and more than
5 times as likely to have a gastrointestinal tract disorder. Respiratory and
neoplastic conditions had the lowest association with developmental delay.
Children with developmental delay were 1 to 2 times as likely to have
these conditions compared with children without developmental delay.
HEALTH CARE USE
Health care use by year of life is shown in Figure 1. Children with developmental delay who also had 1 or more
chronic health conditions had the highest use in all of the categories measured.
The mean number of physician visits, ED visits, and hospitalizations per child
decreased as the age of the child increased. Visits to other practitioners
increased as the age of the child increased. Similar patterns in use were
observed when the data were stratified by sex and race.
|
|
|
|
Health care use by type of health care provider among children with
and without developmental delay and chronic health conditions during the first
5 years of life.
|
|
|
PHYSICIAN VISITS
Children with isolated developmental delay had more visits to a physician
in the first year of life compared with children who had no developmental
delay and no chronic health conditions (P<.001)
(Figure 1A), but fewer than those
children who had chronic health conditions but no developmental delay (P<.001). By the fifth year of life, the number of physician
visits was similar among children with isolated developmental delay and those
without developmental delay who had 1 or more chronic health conditions, and
greater than that for children without developmental delay and without chronic
health conditions (P<.001). The regression model
estimated that children with developmental delay have 1 times as many
physician visits in the first 5 years of life compared with children without
developmental delay (IRR, 1.55; 95% CI, 1.49-1.63).
ED VISITS
In the first 3 years of life, children with chronic health conditions
had a greater mean number of visits to the ED compared with children without
chronic health conditions (P<.001) (Figure 1B). Although the mean number of ED visits per child declined
in all groups, children with developmental delay and chronic health conditions
continued to have a greater mean number of ED visits per child in the fourth
and fifth years of life compared with the other 3 groups (P<.001). In the fifth year of life, children with developmental
delay and chronic health conditions had a mean of 1 visit per child to the
ED, while the other groups had fewer than 0.5 visits per child. Children with
developmental delay had a slight increase in ED use in the first 5 years compared
with children without developmental delay (IRR, 1.09; 95% CI, 1.00-1.18).
HOSPITALIZATIONS
The number of hospitalizations per 100 children was greatest for children
with developmental delay and chronic health conditions, followed by children
with chronic health conditions without developmental delay (Figure 1C). Children without chronic health conditions had few hospitalizations,
a mean of 1 or fewer per 100 children per year of life in the second through
fifth years of life. After the second year of life, only children with developmental
delay and chronic health conditions had a higher mean number of hospitalizations
per 100 children compared with the other 3 groups (P<.01).
Developmental delay was independently associated with an increase in hospitalizations
in the first 5 years of life (IRR, 1.76; 95% CI, 1.42-2.18).
OTHER PRACTITIONER VISITS
The mean number of visits per child per year to other practitioners
was greatest for children with developmental delay (Figure 1D). Children with developmental delay and chronic health
conditions had the greatest mean number of visits to other practitioners in
all years (P<.001). Beginning in the second year
of life, children with developmental delay and no chronic health conditions
had a greater mean number of visits to other practitioners compared with children
without developmental delay (P<.002). Compared
with children without developmental delay, those with developmental delay
had more than 10 times the number of visits to other practitioners (IRR, 10.49;
95% CI, 8.08-13.63).
COMMENT
Our study demonstrates that 1% of the children in this cohort had isolated
developmental delay, nearly one third of those diagnosed as having developmental
delay. Although other studies have estimated the prevalence of young children
with developmental delay,1-4
none, to our knowledge, have assessed the prevalence of isolated developmental
delay.
Children diagnosed as having developmental delay were more likely to
be male. This may reflect a true difference in the prevalence of developmental
delay among boys. However, it may be confounded by a parental propensity to
seek care for boys or health care provider willingness to diagnose developmental
delay in boys. Although we did not identify studies examining sex differences
among young children diagnosed as having developmental delay, several studies
have examined sex differences among children with mental retardation, finding
that use of services for mental retardation is more common in boys.11-13 However, sex differences
have not been noted if standardized testing alone is used to identify children
with mental retardation.12 These differences
may be attributed to behavioral and adaptive differences between boys and
girls. Further studies are needed to determine if these differences seen with
mental retardation hold for children diagnosed as having developmental delay.
Although earlier studies have shown an association between increased
health care use and the presence of developmental delay,1
they have not clarified the role that developmental delay itself plays in
that use apart from chronic health conditions. Our multivariate analysis shows
that developmental delay is associated with increased health care use in the
first 5 years of life in the 4 areas that we assessed. Although we did not
specifically assess the diagnoses associated with the increased number of
visits, the reasons for more frequent physician visits and hospitalizations
among children with developmental delay may include monitoring a child's development,
as well as establishing the diagnosis of developmental delay and assessing
possible causes. The increasing frequency of visits to other practitioners
are likely secondary to evaluation and intervention services. Children with
developmental delay benefit from early intervention with respect to enhanced
development and improved family functioning.14-17
The presence of developmental delay may make management of these children
more challenging for health care providers and families, necessitating greater
health care use. We were unable to address that issue in this study.
The number of visits to other health care providers increases dramatically
after 3 years of age among children with developmental delay. The reason for
this increase may include that the mean age of diagnosis of developmental
delay was almost 34 months, so many children begin receiving services in their
third year of life. Those receiving services before 3 years of age may continue
to receive services after 3 years of age if they continue to meet eligibility
criteria. Additionally, as children get older, there are greater expectations
for what they should be able to do, and new areas of need may manifest as
the children begin to enter a learning environment, prompting referral to
more or different services.18
The rates of health care use in our study are similar to those reported
in other studies.1, 19 In one study,
children younger than 6 years receiving Medicaid or who were uninsured had
a mean hospitalization rate of 15 per 100 children and an ED visit rate of
0.5 per child in the year prior to the study.19
Among children with isolated developmental delay and those without developmental
delay, the rates of hospitalizations were lower than found in this study,
while the rate of ED visits was greater in the first 3 years of life. This
was a cross-sectional study where children were selected based on insurance
status. By selecting children because they had a diagnosis of developmental
delay and including only those who remained continuously enrolled for 12 months
or longer, we overselected children with chronic health conditions and, thus,
greater health care use.
In the National Health Interview Survey, Child Health Supplement (NHIS-CH),
children aged 0 to 17 years with no developmental disabilities had a mean
of 2.8 visits to physicians in the year prior to the survey, and those children
with developmental disabilities had a mean of 6.1 visits to physicians in
the year prior.1 In this same study, children
without disabilities had a mean of 4.0 hospitalizations per 100 children,
and those with disabilities had a mean of 17.8 hospitalizations per 100 children.
Although the mean number of physician visits and hospitalizations in our study
was higher, the NHIS-CH reported a mean across a much broader age range. Other
factors that may account for the differences seen in the NHIS-CH and our study
were that the NHIS-CH was based on parental report, while our numbers are
based on claims submitted. Also the NHIS-CH included a nationwide sample of
children, not just those receiving Medicaid.
There are several limitations to this study. First, because it is based
on claims data, the classification of children with and without developmental
delay and with and without chronic health conditions is only as good as the
diagnoses coded in these data. Claims data are likely to underestimate the
presence of developmental delay and would bias our results against finding
a difference between children with and without developmental delay. However,
we found a 3.2% prevalence of developmental delay among young children continuously
enrolled for at least 1 year from birth, which is consistent with previously
published estimates of developmental delay.1-4
Second, we may have misclassified children as not having developmental delay
when, in fact, they were developmentally delayed. The effect of this type
of misclassification is likely to be very small since the prevalence of developmental
delay is low. Third, we may have missed some use if care was paid for by another
source. Children receiving Supplemental Security Income may have had private
insurance in addition to Medicaid and we were unable to capture use paid for
by other sources. This would lead to an underestimate of use in this study.
The Individuals With Disabilities Education Act Part H (now Part C), the special
education law covering 0- to 3-year-olds, and the public schools provide developmental
services that would not necessarily be reflected in the Medicaid data; however,
these sources often do submit claims to Medicaid. We sought to minimize problems
in ascertainment by including only those children who were continuously enrolled
from birth and only examining the first period of continuous enrollment in
Medicaid. Lastly, this study only assessed use among a continuously enrolled
Medicaid population. These results may not be generalizable to other populations.
Despite these limitations, this study is an important first step in
estimating the prevalence of isolated developmental delay and understanding
the role of developmental delay in young children's health care use. Although
children with developmental delay and chronic health conditions have markedly
increased use, it seems as if chronic health conditions, rather than developmental
delay are responsible for most of that increase; however, developmental delay
may complicate the management of these children. Further studies are needed
to determine if these same patterns of health care use are seen in other non-Medicaid
populations and to assess the role of developmental delay in altering the
management of children with 1 or more chronic health conditions.
| What This Study Adds
Although children with developmental delay are known to have increased
health care use, it is unclear what role developmental delay plays in that
use apart from associated chronic health conditions. Also, the prevalence
of developmental delay without the presence of chronic health conditions is
unknown.
Developmental delay without chronic health conditions affects 1% of
this Medicaid population, and 30% of the children with developmental delay
have no associated chronic health conditions. Developmental delay is associated
with significant health care use, in particular physician visits and services
by other health care providers. Boys are more likely to be diagnosed as having
developmental delay.
|
|
AUTHOR INFORMATION
Accepted for publication November 7, 2001.
This study was supported in part by a Robert Wood Johnson Generalist
Faculty Physician Scholar grant (Dr Christakis).
This study was presented in part at the annual meeting of the Ambulatory
Pediatric Association, Boston, Mass, May 13, 2000.
We thank Ted Grichuhin for programming support. We would like to thank
Heather McPhillips, MD, MPH, and Fred Rivara, MD, MPH, for their editorial
review of the manuscript.
Corresponding author and reprints: Margaret M. Gallaher, MD, MPH,
Children's Medical Services, 2040 S Pacheco, Santa Fe, NM 87505 (e-mail: Mgallaher{at}doh.state.nm.us).
From the Department of Pediatrics and the Child Health Institute, University
of Washington, Seattle. Dr Gallaher is now with Children's Medical Services,
New Mexico Department of Health, Santa Fe.
REFERENCES
| |
1. Boyle CA, Decoufle P, Yeargin-Allsopp M. Prevalence and health impact of developmental disabilities in US children. Pediatrics. 1994;93:399-403.
FREE FULL TEXT
2. Drillien CM, Pickering RM, Drummond MB. Predictive value of screening for different areas of development. Dev Med Child Neurol. 1988;30:294-305.
ISI
| PUBMED
3. Dworkin PH. British and American recommendations for developmental monitoring:
the role of surveillance. Pediatrics. 1989;84:1000-1010.
FREE FULL TEXT
4. Kirby RS, Brewster MA, Canino CU, Pavin M. Early childhood surveillance of developmental disorders by a birth
defects surveillance system: methods, prevalence comparisons, and mortality
patterns. J Dev Behav Pediatr. 1995;16:318-326.
ISI
| PUBMED
5. Disabilities among children aged  17 years—United States,
1991-1992. MMWR Morb Mortal Wkly Rep. 1995;44:609-613.
PUBMED
6. Coiro MJ, Zill N, Bloom B. Health of our nation's children. Vital Health Stat 10. 1994;191:1-61.
7. Kruskal W, Wallis W. Use of ranks in one-criterion variance analysis. J Am Stat Assoc. 1952;47:583-621.
FULL TEXT
|
ISI
8. Rothman K, Greenland S. Modern Epidemiology. 2nd ed. Philadephia, Pa: Lippincott-Raven; 1998.
9. Cornfield J. A statistical problem arising from retrospective studies. In: Neyman J, ed. Proceedings of the Third Berkeley
Symposium on Mathematical Statistics and Probability. Vol 4. Berkeley:
University of California Press; 1956:135-148.
10. Rodriguez G. An improvement to Poisson. Stata Technical Bull. 1993;11:11-14.
11. Katusic SK, Colligan RC, Beard C, et al. Mental retardation in a birth cohort, 1976-1980, Rochester, Minnesota. Am J Ment Retard. 1996;100:335-344.
ISI
| PUBMED
12. Richardson SA, Katz M, Koller H. Sex differences in number of children administratively classified as
mildly mentally retarded: an epidemiologic review. Am J Ment Defic. 1986;91:250-256.
ISI
| PUBMED
13. Murphy C, Yeargin-Allsopp M, Decoufle P, Drews CD. The administrative prevalence of mental retardation in 10-year-old
children in metropolitan Atlanta, 1985 through 1987. Am J Public Health. 1995;85:319-323.
FREE FULL TEXT
14. Guralnick MJ. The next decade of research on the effectiveness of early intervention
[review]. Except Child. 1991;58:174-183.
ISI
| PUBMED
15. Casto G, Mastropieri MA. The efficacy of early intervention programs: a meta-analysis [review]. Except Child. 1986;52:417-424.
ISI
| PUBMED
16. Shonkoff JP, Hauser-Cram P. Early intervention for disabled infants and their families: a quantitative
analysis [review]. Pediatrics. 1987;80:650-658.
FREE FULL TEXT
17. Shonkoff JP, Hauser-Cram P, Frauss MW, Upshur CC. Development of infants with disabilities and their families: implications
for theory and service delivery [review]. Monogr Soc Res Child Dev. 1992;57:v-vii, 1-153.
18. Newacheck PW, Stoddard JJ. Prevalence and impact of multiple childhood chronic illnesses. J Pediatr. 1994;124:40-48.
FULL TEXT
|
ISI
| PUBMED
19. Polivka BJ, Nickel JT, Salsberry PJ, Kuthy R, Shapiro N, Slack C. Hospital and emergency department use by young low-income children. Nurs Res. 2000;49:253-261.
FULL TEXT
|
ISI
| PUBMED
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Definitions and Measurement of Chronic Health Conditions in Childhood: A Systematic Review
van der Lee et al.
JAMA 2007;297:2741-2751.
ABSTRACT
| FULL TEXT
A Long-term, Epidemiological Survey of Outcome and Adjustment of Children with Developmental Disabilities
Leitner et al.
J Child Neurol 2007;22:143-150.
ABSTRACT
Satisfaction With Care and Ease of Using Health Care Services Among Parents of Children With Special Health Care Needs: The Roles of Race/Ethnicity, Insurance, Language, and Adequacy of Family-Centered Care
Ngui and Flores
Pediatrics 2006;117:1184-1196.
ABSTRACT
| FULL TEXT
Trends in Diagnosis Rates for Autism and ADHD at Hospital Discharge in the Context of Other Psychiatric Diagnoses
Mandell et al.
Psychiatr. Serv. 2005;56:56-62.
ABSTRACT
| FULL TEXT
Increased Use of Health Care by Children with Developmental Delay
Warren
AAP Grand Rounds 2002;8:28-29.
FULL TEXT
|
