Trends in Diagnosis Rates for Autism and ADHD at by slappypappy122


									Trends in Diagnosis Rates for Autism
and ADHD at Hospital Discharge in the
Context of Other Psychiatric Diagnoses
David S. Mandell, Sc.D.
William W. Thompson, Ph.D.
Eric S. Weintraub, M.P.H.
Frank DeStefano, M.D., M.P.H.
Michael B. Blank, Ph.D.

Objective: Concerns have been raised over observed increases in the                         ADHD (10,11) and autism (12–14),
number of children who are given a diagnosis of a neurodevelopmental                        few studies have examined trends in
disorder. The goal of this study was to examine trends by age and cal-                      the prevalence of these disorders.
endar year in the diagnosis of two of these disorders, autism and atten-                    Some studies have examined trends
tion-deficit hyperactivity disorder (ADHD), in the context of other psy-                    in treated prevalence. For example,
chiatric disorders in a sample of hospitalized children. Methods: Data                      Kelleher and colleagues (5) found
from the Healthcare Cost and Utilization Project (HCUP) were used for                       that pediatricians identified attention
descriptive analyses of secular trends of diagnosed psychiatric disorders                   problems such as ADHD among 1.4
between 1989 and 2000. Changes over time in rates of diagnosis of                           percent of children in 1979 and 9.2
autism, ADHD, affective disorders, and substance-related disorders                          percent of children in 1996, an in-
were examined and compared. Results: Substance-related disorders                            crease of 657 percent. Data from the
were the most common mental disorders recorded at hospital discharge                        U.S. National Ambulatory Medical
and increased by 39 percent between 1989 and 2000. Affective disorder                       Care Survey suggest a 250 percent
was the next most common diagnosis and increased by 138 percent. Al-                        increase between 1990 and 1998 in
though autism and ADHD were far less common, their diagnosis rates                          the number of children receiving a
nearly quadrupled over the course of the study. Although rates of diag-                     diagnosis of ADHD (15). Zito and
nosis of affective and substance-related disorders generally increased                      colleagues (16) found that children’s
over the lifespan, diagnosis of autism and ADHD followed a very dif-                        ambulatory care discharges for
ferent pattern, with peaks in rates at ages seven and 12. Conclusions: In-                  ADHD increased from 1.9 percent
creases in rates of diagnosis of etiologically unrelated mental disorders                   to 3.6 percent.
suggest that there have been changes in diagnostic practices over time,                        Similar studies have indicated as
increases in community prevalence of these disorders, and increased                         much as a 373 percent increase in the
likelihood of hospitalizations for different mental disorders. (Psychi-                     number of reported cases of autism
atric Services 56:56–62, 2005)                                                              spectrum disorders from 1980 to
                                                                                            1994 (17). California’s Department of
                                                                                            Developmental Services reported an

      here is growing concern about          is associated with environmental               additional 100 percent increase in
      increasing rates of diagnosis of       agents (1–3), improved ascertain-              rates of autism from 1992 to 1997
      autism and attention-deficit           ment (4,5), or overdiagnosis (6–9).            (18). It is unclear whether this ob-
hyperactivity disorder (ADHD) and            Although recent cross-sectional stud-          served change represents a true in-
about whether this observed increase         ies have examined the prevalence of            crease in the community prevalence
                                                                                            or is a function of improved recogni-
                                                                                            tion and diagnosis (4,19–21).
                                                                                               Other studies have shown concur-
Dr. Mandell and Dr. Blank are affiliated with the Center for Mental Health Policy and
Services Research and the Leonard Davis Institute of Health Economics at the Universi-
                                                                                            rent increases in rates of diagnosis of
ty of Pennsylvania in Philadelphia. Dr. Thompson, Mr. Weintraub, and Dr. DeStefano are      other psychiatric disorders among
with the National Immunization Program of the Centers for Disease Control and Pre-          children and adolescents. In their
vention in Atlanta. Send correspondence to Dr. Mandell at CMHPSR, University of Penn-       comprehensive reviews, Rutter and
sylvania, 3535 Market Street, Third Floor, Philadelphia, Pennsylvania 19104 (e-mail,        Smith (22) and Fombonne (23) pro-                                                               vided evidence of increases in rates of
56                                                PSYCHIATRIC SERVICES   ♦ ♦ January 2005 Vol. 56 No. 1
diagnosis of depressive disorders, sui-         lected states that agreed to provide      four periods: 1989 to 1991, 1992 to
cide, and substance use disorders.              data for all payers for inpatient hos-    1994, 1995 to 1997, and 1998 to
Similarly, Achenbach and Howell                 pital discharges. The NIS used a          2000. Differences in trends were ex-
(24) found slight increases in parents’         stratified probability sampling of        amined for each period.
reports of children’s problems. These           community hospitals intended to se-
general increases in pathology paral-           lect 20 percent of each stratum.          Results
lel physicians’ increased recognition           Sampling strata were defined by five      A total of 339,560 hospital discharges
of psychosocial problems (5) and chil-          hospital characteristics: region          were recorded for persons from birth
dren’s increased use of health care             (Northeast, Midwest, West, or             to age 21 years that were associated
services for behavioral and emotional           South), control (government non-          with any psychiatric diagnosis (data
disorders (25,26).                              federal, private not-for-profit, or       not shown). The psychiatric disorder
   It is possible that the community            owned by a private investor), urban-      was the primary diagnosis in 58 per-
prevalence of all these disorders is in-        icity (urban or rural), teaching status   cent of cases. Thirty-three percent of
creasing simultaneously. However, it            (teaching or nonteaching), and bed        psychiatric diagnoses were for drug-
is also possible that changes in                size (small, medium, or large). Each      related disorders; 22 percent of these
parental and social expectations,               year of the NIS contains data for be-     were primary diagnoses. Of the
methods of ascertainment, awareness             tween 800 and 1,000 hospitals and         85,103 discharges (25 percent) associ-
of the signs and symptoms of psychi-            approximately five to seven million       ated with a diagnosis of an affective
atric disorders, and changes in diag-           records. The NIS includes sample          disorder, 76 percent were primary di-
nostic thresholds are responsible for           weights that provide both state- and      agnoses. Of the 30,317 discharges (9
the observed increase in rates of diag-         national-level estimates.                 percent) associated with a diagnosis
nosis of these disorders. In any case, it                                                 of ADHD, 24 percent were primary
is critical that these changes be exam-         Sample                                    diagnoses. Of the 3,736 discharges (1
ined in the context of other—pre-               This study included all hospital dis-     percent) associated with an autism di-
sumably unrelated—disorders, so                 charges from calendar years 1989 to       agnosis, 23 percent were primary di-
that patterns can be compared. To               2000 in the HCUP NIS for individu-        agnoses.
our knowledge, no study has exam-               als from birth to age 21 years. Indi-       An overall increase was noted in
ined the changing rates of these dis-           viduals could be included in the sam-     the hospital discharge rate for each
orders over time in the same data set.          ple more than once if they had multi-     diagnosis. Diagnoses of alcohol and
The goal of the study reported here             ple discharges.                           other drug use disorders were most
was to examine trends by age and cal-                                                     common, increasing from 125 per
endar year in the diagnosis of autism           Measures                                  100,000 discharges (95 percent confi-
spectrum disorders and ADHD at                  Psychiatric disorders were coded by       dence interval [CI]=114 to 136 per
hospital discharge in the context of            using the International Classification    100,000) in 1989 to 174 per 100,000
hospital discharges for other psychi-           of Diseases, 9th Edition (ICD-9) (27).    (CI=160 to 189 per 100,000) in 2000,
atric disorders.                                Disorders were associated with a hos-     followed by major affective disorders,
                                                pital discharge if they were coded in     which increased from 65 per 100,000
Methods                                         any position for that discharge. Co-      (CI=60 to 71 per 100,000) to 155 per
Data from the Healthcare Cost and               morbid conditions were double-            100,000 (CI=134 to 177 per 100,000);
Utilization Project (HCUP) were                 counted for all analyses. Autism spec-    ADHD, which increased from 11 per
used to examine secular trends in di-           trum disorders were coded for any         100,000 (CI=9 to 14 per 100,000) to
agnosed psychiatric disorders be-               discharge with a 299.0 diagnostic         53 per 100,000 (CI=45 to 61 per
tween 1989 and 2000. The HCUP                   code, ADHD for any discharge with         100,000); and autism, which in-
Nationwide Inpatient Sample (NIS)               codes 314.00 to 314.99, major affec-      creased from 1.2 per 100,000 (CI=1
approximates a 20 percent sample of             tive disorders for any discharge with     to 1.4 per 100,000) to 5.5 per 100,000
U.S. community hospitals. The Amer-             codes 296.00 to 296.99, and sub-          (CI=4.4 to 6.5 per 100,000). Although
ican Hospital Association defines               stance use disorder for any discharge     the discharge rates associated with di-
community hospitals as “all nonfeder-           with codes 303.00 to 305.99.              agnoses of alcohol and other drug use
al, short-term, general, and other spe-                                                   disorders dipped in the early 1990s,
cialty hospitals, excluding hospital            Analyses                                  the trend for these and affective dis-
units of hospital institutions.” Special-       For each set of disorders, three rates    orders were relatively parallel over
ty hospitals, such as obstetric-gyne-           were calculated. First, the rate of       the study period. Discharge rates for
cology, ear-nose-throat, short-term             hospital discharges associated with       autism and ADHD evidenced a
rehabilitation, orthopedic, and pedi-           each disorder was calculated for          greater relative increase over the
atric hospitals, were included, and             each calendar year as a function of       study period.
long-term hospitals, psychiatric hos-           the total number of hospital dis-           Hospital discharge rates for each
pitals, and chemical dependency                 charges for that year. Second, aver-      diagnosis as a function of age are
treatment facilities were not included          age rates were calculated across all      shown in Table 1. Rates of diagnosis
in the survey.                                  years of the study period by age. Fi-     of alcohol and drug use disorder were
   The NIS includes data from se-               nally, the sample was divided into        close to zero among children up to
PSYCHIATRIC SERVICES   ♦ ♦ January 2005 Vol. 56 No. 1                                          57
Table 1                                                                                     the age of 12 years, reaching a peak of
                                                                                            466 per 100,000 for 17-year-olds and
Average annual diagnosis rates per 100,000 discharges, by age and disorder
                                                                                            then peaking again at 529 per 100,000
              Attention-deficit                                                             for 21-year-olds. Rates of diagnosis of
              hyperactivity                         Affective        Alcohol or drug        affective disorders began to increase
Age (years)   disorder              Autism          disorders        dependence             among younger children and to in-
                                                                                            crease more gradually to a peak for
 0              0                   0                 1                8
 1              1                   0                 0                1
                                                                                            17-year-olds (319 per 100,000) and
 2              4                   2                 0                1                    decrease for 18-year-olds. Diagnoses
 3              5                   4                 0                0                    of ADHD appeared among children
 4             13                   5                 2                0                    around four years of age and in-
 5             24                   5                 5                0                    creased gradually, peaking among
 6             35                   4                 9                0
 7             67                   6                20                0
                                                                                            seven-year-olds (67 per 100,000) and
 8             54                   4                25                6                    12-year-olds (90 per 100,000) and de-
 9             52                   4                29                1                    creasing for older individuals. Diag-
10             55                   3                37                1                    noses of autism increased more
11             58                   3                50                3                    sharply among young children, reach-
12             90                   4               120               28
13             70                   3               147               52
                                                                                            ing a rate of 4.2 per 100,000 among
14             68                   3               217              128                    three-year-olds and peaking at 5.9 per
15             63                   3               266              215                    100,000 among seven-year-olds. A
16             51                   3               273              293                    secondary peak was observed among
17             44                   3               319              466                    12-year-olds (4.2 per 100,000).
18             17                   2               199              391
19             11                   2               186              425
                                                                                               Average annual discharge rates as-
20              8                   2               185              459                    sociated with diagnoses of alcohol
21              7                   2               193              529                    and other substance use disorders
                                                                                            are shown in Table 2. The trends for
                                                                                            all four periods follow a relatively
                                                                                            similar pattern, with diagnoses be-
                                                                                            ginning among 12-year-olds and in-
                                                                                            creasing steadily until age 21. A sec-
                                                                                            ondary peak was noted among 17-
                                                                                            year-olds. The relatively small differ-
                                                                                            ences across each period for children
Table 2
                                                                                            and adolescents suggest that the
                                                                                            changes observed in Table 1 are a
Average annual rates of diagnosis of alcohol and drug dependence per 100,000                function of increasing discharge
discharges, by three-year cohort and age at discharge                                       rates among young adults. For exam-
               Three-year cohort                                                            ple, among 21-year-olds, there were
                                                                                            680 per 100,000 diagnoses in the pe-
Age (years)    1989 to 1991        1992 to 1994      1995 to1997         1998 to 2000       riod 1998 to 2000, compared with
                                                                                            385 per 100,000 in 1989 to 1991.
 0               8                   6                 6                  11                   Average annual discharge rates for
 1               1                   1                 0                   1
 2               2                   1                 1                   2                major affective disorders are shown in
 3               0                   0                 0                   0                Table 3. Children begin to have hos-
 4               0                   0                 0                   0                pital discharges with an associated di-
 5               0                   0                 0                   0                agnosis of major affective disorder
 6               0                   0                 0                   0                around the age of five years. Although
 7               1                   0                 0                   0
 8               0                   0                 0                  21                all four groups followed a similar pat-
 9               0                   1                 1                   1                tern across the age spectrum, in-
10               1                   1                 1                   1                creased differentiation was noted
11               2                   2                 4                   3                over time, with greater increases in
12              28                  23                35                  28                more recent periods. Discharge rates
13              51                  46                63                  47
14             135                 110               147                 120                associated with affective disorders
15             214                 180               253                 215                peaked at age 17, at 409 per 100,000
16             283                 242               340                 308                in the period 1998 to 2000 and at 248
17             475                 366               503                 521                per 100,000 in the period 1989 to
18             280                 315               465                 504                1991. In all four periods, a decrease
19             289                 343               482                 586
20             322                 387               510                 617                after age 17 was noted.
21             385                 469               582                 680                   Average annual discharge rates as-
                                                                                            sociated with a diagnosis of ADHD
58                                                PSYCHIATRIC SERVICES   ♦ ♦ January 2005 Vol. 56 No. 1
are shown in Table 4. Discharges as-            Table 3
sociated with this diagnosis can be ob-
                                                Average annual rates of diagnosis of affective disorders per 100,000 discharges, by
served among children as young as
                                                three-year cohort and age at discharge
two years of age, with a sharp rise un-
til age seven; more recent periods                                 Three-year cohort
showed a greater increase. The dis-
charge rate for seven-year-olds was 27          Age (years)        1989 to 1991        1992 to 1994   1995 to 1997    1998 to 2000
per 100,000 discharges in the period
                                                 0                    1                  1              1               1
1989 to 1991 and 96 per 100,000 in               1                    0                  0              0               0
the period 1998 to 2000. Rates de-               2                    1                  0              0               1
clined among children aged seven to              3                    0                  1              0               0
eight years, increased again among               4                    1                  2              1               2
children aged 11 to 12, dropped again            5                    3                  5              5               6
                                                 6                    4                  9              9              12
among children aged 12 to 13, and                7                    8                 18             21              33
then fell more gradually to a low                8                   11                 22             25              43
among 21-year-olds. This pattern was             9                   15                 27             28              46
more pronounced in recent periods.              10                   19                 36             36              58
   Average annual discharge rates as-           11                   29                 42             51              78
                                                12                   71                102            126             180
sociated with autism are shown in               13                  100                135            154             199
Table 5. The trends by calendar year            14                  154                206            225             281
as well as by age suggest a similar pat-        15                  192                246            282             346
tern to that of ADHD. Discharge                 16                  191                247            294             362
rates associated with autism increased          17                  248                283            334             409
                                                18                  128                178            228             262
sharply among children aged one to              19                  113                163            210             257
three years and again among children            20                  123                164            205             246
aged six to seven years, with greater           21                  128                175            220             249
increases in more recent periods.
Among seven-year-olds, 11 per
100,000 children received a diagnosis
of an autism spectrum disorder in the
period 1998 to 2000, compared with
three per 100,000 in 1989 to 1991.
Rates then dropped among children
aged seven to 11, with greater de-              Table 4
creases in more recent years. Another
increase was seen among children                Average annual rates of diagnosis of attention-deficit hyperactivity disorder per
aged 11 to 12 years, with another               100,000 discharges, by three-year cohort and age at discharge
drop from age 12 to 13. Among 12-                                  Three-year cohort
year-olds, eight per 100,000 children
were given a diagnosis in the period            Age (years)        1989 to 1991        1992 to 1994   1995 to 1997    1998 to 2000
1998 to 2000, compared with two per
                                                 0                   0                  0               0               0
100,000 in the period 1989 to 1991.
                                                 1                   1                  1               1               0
By young adulthood, there was less               2                   3                  4               5               4
differentiation by period in the pro-            3                   3                  6               7               5
portion of hospital discharges.                  4                   8                 14              16              13
                                                 5                  12                 26              27              30
                                                 6                  17                 34              45              43
Discussion                                       7                  27                 59              84              96
This study provided evidence of in-              8                  21                 48              69              77
creasing hospital discharge rates for            9                  21                 44              64              82
several psychiatric disorders between           10                  22                 45              67              87
1989 and 2000. Although substance-              11                  26                 44              70              92
                                                12                  36                 67             109             149
related and affective disorders were
                                                13                  30                 61              87             101
more common, the increase was                   14                  27                 61              86              97
more pronounced for ADHD, which                 15                  25                 52              81              92
evidenced a 381 percent increase                16                  20                 40              69              77
over the study period, and autism,              17                  15                 31              55              75
                                                18                   5                 11              22              31
which evidenced a 358 percent in-               19                   2                  6              15              20
crease. Increases were observed in              20                   2                  5              11              15
successive periods for all disorders.           21                   2                  4              10              13
Diagnoses of ADHD and autism
PSYCHIATRIC SERVICES   ♦ ♦ January 2005 Vol. 56 No. 1                                            59
Table 5                                                                                   changing practice patterns. The fact
Average annual rates of diagnosis of autism per 100,000 discharges, by three-year         that a greater increase in diagnoses of
cohort and age at discharge                                                               developmental disorders among
                                                                                          younger children was observed in the
            Three-year cohort                                                             more recent periods suggests the pos-
                                                                                          sibility of increased community
Age         1989 to 1991        1992 to 1994      1995 to 1997          1998 to 2000
                                                                                          prevalence. A third possibility is that
 0           1                   0                 0                     0                children with developmental disor-
 1           0                   0                 0                     1                ders are more likely to be hospitalized
 2           1                   1                 2                     4                than are other children.
 3           2                   3                 5                     8
 4           2                   3                 6                     8
 5           1                   4                 5                     8
                                                                                          Increased ascertainment
 6           2                   3                 5                     7                A MEDLINE search using the term
 7           3                   4                 5                    11                “attention deficit disorder” returned
 8           1                   2                 4                     8                185 articles in 1989 and 541 in 2000.
 9           1                   2                 3                     7                A search that used the term “autistic
10           2                   3                 3                     6
11           1                   2                 3                     4
                                                                                          disorder” returned 216 articles in
12           2                   2                 4                     8                1989 and 359 in 2000. Increased
13           1                   2                 3                     5                awareness was not limited to the
14           2                   2                 3                     5                medical literature. A NEXIS search
15           1                   3                 3                     6                of all U.S. news outlets found 25 arti-
16           1                   2                 3                     5
17           1                   2                 3                     5
                                                                                          cles on ADHD and autism in 1988
18           1                   2                 2                     4                and 838 articles in 2000. Popular
19           1                   2                 2                     3                awareness may have consequently in-
20           1                   1                 2                     3                creased, fueling parental concern
21           1                   1                 2                     3                and, in turn, changing physicians’ be-
                                                                                          havior (31). Concurrently, treatments
                                                                                          for developmental disorders have be-
                                                                                          come more widely accepted and
showed peaks among seven- and 12-          but it appears that the underestima-           used. More children with a diagnosis
year-olds. In contrast, substance-re-      tion is not dissimilar across disorders.       of ADHD (32) as well as younger
lated and affective disorders showed          One might expect that hospitaliza-          children with this diagnosis (33) are
gradual increases in discharge rates,      tion rates would be higher among               treated with stimulants. Likewise,
peaking among 17-year-olds.                older children, especially in the case         children with autism are increasingly
   As expected, the rates of diagnoses     of chronic conditions. This was the            treated with stimulants and neurolep-
among individuals being discharged         general pattern for affective and sub-         tics (34). The increasing popularity of
from a hospital were lower than com-       stance-related disorders. However,             these treatment options may increase
munity prevalence rates. Survey data       rates of diagnosis of autism and               the likelihood that providers will as-
suggest that between 20 and 40 per-        ADHD were highest among seven-                 sign a related diagnosis.
cent of adolescents have been drunk        and 12-year-olds, with rates drop-                Another possible reason for chang-
in the past 30 days, and between 4         ping between ages seven and 12 and             ing ascertainment is the expansion of
and 12 percent have used hallucino-        again after age 12. It is unlikely that        Supplemental Security Income (SSI)
gens, narcotics, or cocaine in the past    related symptoms decreased be-                 and the Individuals With Disabilities
30 days (28). Although these percent-      tween these ages. Rather, it is likely         Education Act (IDEA), which may
ages are overestimations of the pro-       that certain events precipitate the as-        have provided incentive for physicians
portion of youths who would qualify        sociation of a diagnosis of autism or          to diagnose ADHD and autism. The
for a diagnosis of substance abuse or      ADHD and a hospital discharge. For             number of children insured through
dependence, the numbers suggest            example, the peaks among seven-                SSI increased by more than 400 per-
that the maximum rate of .5 percent        and 12-year-olds suggest that transi-          cent between 1988 and 1997 (35), in
found in this study is an underestima-     tions into elementary school or into           large part because of the Social Secu-
tion. Likewise, the highest estimate of    middle school are associated with in-          rity Administration’s broadened defi-
affective disorders in this population     creased rates of diagnosis.                    nition of disability to include mild
was .4 percent, whereas estimates of          The finding that hospital diagnoses         mental retardation and mental illness.
affective disorders among children         of mental disorders are increasing                Concurrently, the number of chil-
have been between 3 and 8 percent          across the age span in more recent             dren served by IDEA grew from 4.5
(29,30). The prevalence of ADHD is         years has several possible interpreta-         to 6.3 million between 1988 and 2000
between 3 and 5 percent (10,11) and        tions. Increased rates of diagnosis of         (36). The categories of both autism
of autism is between .2 and .4 percent     all disorders over time suggest in-            and “other health impairment,”
(14). These data provide underesti-        creased ascertainment, perhaps due             which includes ADHD, grew by more
mations of community prevalence,           to increased awareness of disorders or         than 1,000 percent during this time.
60                                              PSYCHIATRIC SERVICES   ♦ ♦ January 2005 Vol. 56 No. 1
In 1990, IDEA was expanded to in-               sion for asthma (50) and influenza (51).     treatment facilities, so individuals
clude early intervention services for              Gallaher and colleagues (52) found        with alcohol and drug dependence
children aged three to five years (36).         that having a developmental delay was        are probably underrepresented. The
School systems were required to im-             independently associated with hospi-         number of such specialty hospital fa-
plement new systems for screening               talization, even after other health con-     cilities have also been declining in re-
and diagnosis, which may have in-               ditions had been controlled for. They        cent years, and few children are ad-
creased the number of children re-              also found that children with develop-       mitted to these types of facilities, so it
ceiving a diagnosis.                            mental delays were twice as likely as        is difficult to ascertain exactly what
                                                other children to have other chronic         bias the exclusion of the facilities may
Increase in community prevalence                health conditions. Children with both        have on our results.
Byrd and colleagues (18) suggest that           developmental delay and another
changes in diagnostic practices can-            chronic health condition were much           Conclusions
not explain the rise in diagnosis of            more likely to be hospitalized than          Regardless of the reason for the ob-
autism. Several etiologic pathways for          other children, and, unlike the effect       served increase, the results of this
this increase have been posited. One            of having a chronic health condition         study suggest that increasing rates of
hypothesis is an increase in exposure           only, this difference persisted over the     diagnosis of autism and ADHD at
to environmental toxins, such as mer-           first five years of life. Increased proba-   hospital discharge are not occurring
cury (3,37,38). A second is an in-              bility of hospitalization among chil-        in isolation but, rather, in the context
crease in viral loads associated with           dren with developmental disabilities         of increasing rates of diagnosis of a
measles, mumps, and rubella (MMR)               may partially account for trends ob-         number of disorders. Although it is
vaccines. The association between in-           served in this study.                        possible that the community preva-
creasing rates of both MMR vaccina-                                                          lence of various psychiatric and de-
tion and autism has been noted (17),            Limitations                                  velopmental disorders is increasing, it
but evidence of causality is limited            Some limitations of this study should        is important to consider variables
(4,39). Recent studies indicated no             be mentioned. As with any adminis-           such as changes in relevant policies,
association between MMR vaccina-                trative data source, the HCUP is sub-        increased awareness, and changes in
tion and autism (40–42), and limited            ject to coding errors (53). If these er-     practice patterns that may also be
data are available on the risk of neu-          rors were differentially biased by dis-      causally related to these increases. ♦
rodevelopmental disorders associated            order, or error rates changed over
with thimerosal exposure (41). A sec-           time, it would bias comparisons with-        References
ond possibility relates to the increas-         in or among diagnoses. Second, it is         1. Halsey N, Hyman S: Measles-mumps-
ing survival rate of low-birth-weight           possible that although individuals              rubella vaccine and autistic spectrum disor-
                                                                                                der: report from the New Challenges in
and otherwise compromised babies,               with psychiatric disorders presented            Childhood Immunizations Conference
who are at increased risk of develop-           at hospitals in the data set, psychiatric       convened in Oak Brook, Illinois, June
mental disorders (43). From 1989 to             diagnoses were not assigned. It may             12–13, 2000. Pediatrics 107:E84, 2001
1997, rates of mortality among                  be that when presenting problems are         2. Hornig M, Lipkin W: Infectious and im-
preterm infants decreased by more               more directly related to the mental             mune factors in the pathogenesis of neu-
                                                                                                rodevelopmental disorders: epidemiology,
than 30 percent (44). However, it is            disorder—for example, intoxication              hypotheses, and animal models. Mental
unlikely that the increase in the num-          and substance use or self-injury and            Retardation and Developmental Disabili-
ber of low-birth-weight babies alone            depression—the disorder is more                 ties Research Reviews 7:200–210, 2001
could account for a large change in             likely to be coded. This dynamic             3. Myers G, Davidson P: Does methylmer-
rates of developmental disorders.               might lead to underreporting of rates           cury have a role in causing developmental
                                                                                                disabilities in children? Environmental
                                                of diagnosis of ADHD and autism rel-            Health Perspectives 108:413–420, 2000
Increase in prevalence                          ative to other disorders. A related
                                                                                             4. Fombonne E, Chakrabarti S: No evidence
of hospitalization                              limitation is that for approximately            for a new variant of measles-mumps-rubel-
A number of studies have shown that             three-quarters of individuals with a            la-induced autism. Pediatrics 108:E58,
children with developmental disorders           diagnosis of an affective disorder,             2001
use more health care services than oth-         ADHD, or an autism spectrum disor-           5. Kelleher K, McInerny T, Gardner W, et al:
er children (45,46). Although hospital-         der, the diagnosis was secondary to             Increasing identification of psychosocial
                                                                                                problems: 1979–1996. Pediatrics 105:
ization rates for the general population        some other morbidity. Primary diag-             1313–1321, 2000
remained constant from 1983 to 1991,            noses were not examined in these cas-
                                                                                             6. Boland R, Diaz S, Lamdan R, et al: Over-
rates of hospitalizations for persons           es and may have been differentially             diagnosis of depression in the general hos-
with developmental disabilities in-             associated with the assigned second-            pital. General Hospital Psychiatry 18:28–
creased by 56 percent (47). This                ary diagnosis. Third, because this              35, 1996
change could not be accounted for by            study used only inpatient hospital           7. Goldman L, Genel M, Bezman R, et al: Di-
an increase in the number of persons            data from specific types of hospitals,          agnosis and treatment of attention-
                                                                                                deficit/hyperactivity disorder in children
with developmental disabilities and oc-         the results may not be generalizable            and adolescents: Council on Scientific Af-
curred at the same time as a reduction          to other health care settings. Fourth,          fairs, American Medical Association. JAMA
in hospital admissions for pediatric in-        the hospitalization data do not in-             279:1100–1107, 1998
jury (48,49) and stable rates of admis-         clude long-term psychiatric and drug         8. LeFever G, Dawson K, Morrow A: The ex-

PSYCHIATRIC SERVICES   ♦ ♦ January 2005 Vol. 56 No. 1                                                     61
     tent of drug therapy for attention deficit-    24. Achenbach T, Howell C: Are American                 troversy laid to rest? CNS Drugs 15:831–
     hyperactivity disorder among children in           children’s problems getting worse? A 13-            837, 2001
     public schools. American Journal of Public         year comparison. Journal of the American
     Health 89:1359–1364, 1999                          Academy of Child and Adolescent Psychia-         40. Madsen K, Hviid A, Vestergaard M, et al: A
                                                        try 32:1145–1154, 1993                               population-based study of measles, mumps,
 9. Safer D: An outpatient/inpatient compari-                                                                and rubella vaccination and autism. New
    son of child psychiatric diagnosis. American    25. Manderscheid R, Henderson M: Mental                  England Journal of Medicine 34:1477–
    Journal of Orthopsychiatry 65:298–303,              Health, United States, 2000. Rockville, Md,          1482, 2002
    1995                                                US Department of Health and Human Ser-
                                                        vices, 2001                                      41. Stratton K, Gable A, McCormick M:
10. Connor D: Preschool attention deficit hy-                                                                Thimerosal-Containing Vaccines and Neu-
    peractivity disorder: a review of prevalence,   26. Olfson M, Marcus S, Druss B, et al: Na-              rodevelopmental Disorders. Washington,
    diagnosis, neurobiology, and stimulant              tional trends in the outpatient treatment of         DC, National Academy Press, 2001
    treatment. Journal of Developmental and             depression. JAMA 287:203–209, 2002
    Behavioral Pediatrics 23:S1–S9, 2002                                                                 42. Taylor B, Miller E, Farrington C, et al:
                                                    27. Medicode: International Classification of            Autism and measles, mumps, and rubella
11. Scahill L, Schwab-Stone M: Epidemiology             Diseases, Ninth Edition. Salt Lake City,             vaccine: no epidemiological evidence for a
    of ADHD in school-age children. Child and           Utah, Med-Index Publications, 1994                   causal association. Lancet 353:2026–2029,
    Adolescent Psychiatric Clinics of North                                                                  1999
    America 9:541–555, 2000                         28. Bauman A, Phongsavan P: Epidemiology of
                                                        substance use in adolescence: prevalence,        43. Lorenz J, Wooliever D, Jetton J, et al: A
12. Bertrand J, Mars A, Boyle C, et al: Preva-          trends, and policy implications. Drug and            quantitative review of mortality and devel-
    lence of autism in a United States popula-          Alcohol Dependence 55:187–207, 1999                  opmental disability in extremely premature
    tion: the Brick Township, New Jersey, in-                                                                newborns. Archives of Pediatrics and Ado-
    vestigation. Pediatrics 108:1155–1161,          29. Geller B, Luby J: Child and adolescent               lescent Medicine 152:425–435, 1998
    2001                                                bipolar disorder: a review of the past 10
                                                        years. Journal of the American Academy of        44. Demissie K, Rhoads G, Ananth C, et al:
13. Fombonne E: The epidemiology of autism:             Child and Adolescent Psychiatry 36:1168–             Trends in preterm birth and neonatal mor-
    a review. Psychological Medicine 29:769–            1176, 1997                                           tality among blacks and whites in the Unit-
    786, 1999                                                                                                ed States from 1989 to 1997. American
                                                    30. Robb A: Bipolar disorder in children and             Journal of Epidemiology 154:307–315,
14. Yeargin-Allsopp M, Rice C, Karapurkar T,            adolescents. Current Opinion in Pediatrics           2001
    et al: Prevalence of autism in a US metro-          11:317–322, 1999
    politan area. JAMA 289:49–55, 2003                                                                   45. Boyle C, Decoufle P, Yeargin-Allsopp M:
                                                    31. Dulcan M, Costello E, Costello A, et al:             Prevalence and health impact of develop-
15. Robison L, Skaer T, Sclar D, et al: Is atten-       The pediatrician as gatekeeper to mental             mental disabilities in US children. Pedi-
    tion deficit hyperactivity disorder increas-        healthcare for children: do parents’ con-            atrics 93:399–403, 1994
    ing among girls in the US? Trends in diag-          cerns open the gate? Journal of the Ameri-
    nosis and the prescribing of stimulants.            can Academy of Child and Adolescent Psy-         46. Drillien C, Pickering R, Drummond M:
    CNS Drugs 16:129–137, 2002                          chiatry 29:453–458, 1990                             Predictive value of screening for different
16. Zito J, Safer D, dosReis S, et al: Psy-                                                                  areas of development. Developmental
                                                    32. Shatin D, Drinkard C: Ambulatory use of              Medicine and Child Neurology 30:294–
    chotherapeutic medication patterns for
                                                        psychotropics by employer-insured chil-              305, 1988
    youths with attention-deficit/hyperactivity
                                                        dren and adolescents in a national managed
    disorder. Archives of Pediatrics and Adoles-                                                         47. Walsh K, Kastner T, Criscione T: Charac-
                                                        care organization. Ambulatory Pediatrics 2:
    cent Medicine 153:1257–1263, 1999                                                                        teristics of hospitalizations for people with
                                                        111–119, 2002
17. Dales L, Hammer S, Smith N: Time trends                                                                  developmental disabilities: utilization,
                                                    33. Rappley M, Eneli I, Mullan P, et al: Pat-            costs, and impact of care coordination.
    in autism and in MMR immunization cov-
                                                        terns of psychotropic medication use in              American Journal of Mental Retardation
    erage in California. JAMA 285:1183–1185,
                                                        very young children with attention-deficit           101:505–520, 1997
                                                        hyperactivity disorder. Journal of Develop-
18. Byrd R: Autistic Spectrum Disorders:                mental and Behavioral Pediatrics 23:23–30,       48. Durbin D, Schwarz D, Localio A, et al:
    Changes in the California Caseload: An              2002                                                 Trends in incidence of pediatric injury hos-
    Update: 1999–2002. Sacramento, Calif,                                                                    pitalizations in Pennsylvania. American
    California Health and Human Services            34. Volkmar FR: Pharmacological interven-                Journal of Public Health 90:1782–1784,
    Agency, 2003                                        tions in autism: theoretical and practical is-       2000
                                                        sues. Journal of Clinical Child Psychology
19. Fombonne E: Is there an epidemic of                 30:80–87, 2001                                   49. Thurman D, Guerrero J: Trends in hospi-
    autism? Pediatrics 107:411–413, 2001                                                                     talization associated with traumatic brain
                                                    35. Loprest P: Children Receiving SSI. Wash-             injury. JAMA 282:954–957, 1999
20. Taylor B, Miller E, Lingam R, et al:                ington, DC, Urban Institute, 1997
    Measles, mumps, and rubella vaccination                                                              50. Russo M, McConnochie K, McBride J, et
    and bowel problems or developmental re-         36. US Department of Education: Twenty-                  al: Increase in admission threshold explains
    gression in children with autism: popula-           Third Annual Report to Congress on the               stable asthma hospitalization rates. Pedi-
    tion study. British Medical Journal 324:            Implementation of the Individuals With               atrics 104:454–462, 1999
    393–396, 2002                                       Disabilities Education Act. Jessup, Md, Bu-
                                                        reau of Special Education, 2001                  51. Schoendorf K, Adams W, Kiely J, et al: Na-
21. Wing L, Potter D: The epidemiology of                                                                    tional trends in Haemophilus influenzae
    autistic spectrum disorders: is the preva-      37. Bigham M, Copes R, Srour L: Exposure to              meningitis mortality and hospitalization
    lence rising? Mental Retardation and De-            themerosal in vaccines used in Canadian in-          among children, 1980 through 1991. Pedi-
    velopmental Disabilities Research Reviews           fant immunization programs, with respect             atrics 93:663–668, 1994
    8:151–161, 2002                                     to risk of neurodevelopmental disorders.
                                                        Canada Communicable Disease Report               52. Gallaher M, Christakis D, Connell F:
22. Rutter M, Smith D: Psychological Disor-             28:69–80, 2002                                       Health care use by children diagnosed as
    ders in Young People: Time Trends and                                                                    having developmental delay. Archives of
    Their Causes. Chichester, England, Wiley,       38. Stein J, Schettler T, Wallinga D, et al: In          Pediatrics and Adolescent Medicine 156:
    1995                                                harm’s way: toxic threats to child develop-          246–251, 2002
                                                        ment. Journal of Developmental and Be-
23. Fombonne E: Increased rates of psychoso-            havioral Pediatrics 23:S13–S22, 2002             53. Berthelsen C: Evaluation of coding data
    cial disorders in youth. European Archives                                                               quality of the HCUP National Inpatient
    of Psychiatry and Clinical Neuroscience         39. DeStefano F, Chen R: Autism and                      Sample. Topics in Health Information
    248:14–21, 1998                                     measles-mumps-rubella vaccination: con-              Management 21:10–23, 2000

62                                                        PSYCHIATRIC SERVICES      ♦ ♦ January 2005 Vol. 56 No. 1

To top