[Federal Register Volume 63, Number 137 (Friday, July 17, 1998)]
[Notices]
[Pages 38661-38665]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 98-19039]


-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Substance Abuse and Mental Health Services Administration


Children With Serious Emotional Disturbance; Estimation 
Methodology

AGENCY: Center for Mental Health Services, Substance Abuse and Mental 
Health Services Administration, HHS.

ACTION: Final notice.

-----------------------------------------------------------------------

SUMMARY: This notice describes the final methodology to identify and 
estimate the number of children with a serious emotional disturbance 
(SED) within each State. This notice is being published as part of the 
requirements of Public Law 102-321, the ADAMHA Reorganization Act of 
1992.

EFFECTIVE DATE: October 1, 1998.

Background

    Public Law 102-321, the ADAMHA Reorganization Act of 1992, amended 
the Public Health Service Act and created the Substance Abuse and 
Mental Health Services Administration (SAMHSA). The Center for Mental 
Health Services (CMHS) was established within SAMHSA to coordinate 
Federal efforts in the prevention and treatment of mental illness, and 
the promotion of mental health. Title II of Public Law 102-321 
establishes a Block Grant for Community Mental Health Services, 
administered by CMHS, that permits the allocation of funds to States 
for the provision of community mental health services for children with 
a SED and adults with a serious mental illness (SMI). Public Law 102-
321 stipulates that States estimate the incidence (number of new cases) 
and prevalence (total number of cases in a year) of individuals with 
either SED or SMI in their applications for block grant funds.

[[Page 38662]]

    As part of the process of implementing this new block grant, 
definitions of the terms ``children with a serious emotional 
disturbance'' and ``adults with a serious mental illness'' were 
announced on May 20, 1993, in Federal Register Notice, Volume 58, No. 
96, p. 29422. Subsequently, a group of technical experts was convened 
by CMHS to develop an estimation methodology to ``operationalize'' the 
key concepts in the definition of children with SED. A similar group 
prepared an estimation methodology for adults with a SMI (March 28, 
1997, Federal Register Notice, Volume 62, No. 60 p.14928).

Summary of Comments

    This document reflects a thorough review and analysis of comments 
received in response to an earlier notice published in the Federal 
Register, on October 6, 1997. Ten letters expressing either support or 
concern regarding the proposed methodology were received by the close 
of the public comment period. Those expressing support praised the 
effort of the CMHS team of technical experts to develop reliable State 
estimates for the number of children with SED. Comments expressing 
concern generally noted limitations similar to those identified by the 
team of technical experts in the original October 6, 1997, Federal 
Register notice. These limitations included the exclusion of children 
from birth to age 8 and the exclusion of variables such as ethnicity 
and geographical location. Additionally, concerns were raised about 
whether the proposed methodology represented prevalence rates more 
precisely than State surveys or local data collection efforts.
    Before addressing the comments, CMHS extends appreciation to 
representatives from Atlantic County, New Jersey, and the University of 
Texas Medical Branch at Galveston for directing attention to errors 
made in Table 3--1995 Estimates of Children and Adolescents with SED by 
State. The New Jersey upper limit for less-impaired children should 
read 102,594, and the Utah upper limit estimate should read 38,399. 
These corrections to Table 3 have been made and will be reflected in 
all subsequent publications.

Purpose of the Methodology

    Although several comments indicated satisfaction with the 
estimation methodology, several others requested that CMHS clarify 
appropriate use of the methodology. In response, CMHS emphasizes that 
the methodology for children and adolescents with SED was developed 
specifically for States to use in the areas of planning and program 
development. Since it is obvious that resources for this population of 
children are inadequate in relation to need, States should continue to 
set priorities to assure the most cost-effective use of all available 
resources. Inclusion or exclusion of any individual based on this 
methodology is not intended to either confer or deny eligibility for 
any other service or benefit at the Federal, State, or local level.

Estimation Methods

    Some comments suggested that surveys and other State-specific or 
local data would provide more precise estimations than the proposed 
methodology. CMHS understands this concern. However, a group of 
technical experts established by CMHS determined that the most valid 
method to estimate the prevalence of SED was to examine findings from 
extant community epidemiological studies that used a structured 
diagnostic interview connected to the DSM-III or DMS-III-R system. The 
group of technical experts thoroughly searched for studies that met 
this criteria and incorporated findings from all of the studies in its 
report. CMHS recognizes the value of local or statewide surveys but 
continues to support the view that the most valid estimates can be 
derived from community epidemiological studies that have used a 
structured diagnostic interview. CMHS will support the use of State 
data if they are based on community epidemiological studies that 
include a standardized diagnostic interview that is linked with the DSM 
system and that also includes a measurement of functional impairment.
    Concerns were also raised that the singular use of poverty as an 
adjustment to prevalence rates was based on convenience. This is not 
the case and the October 6, 1997, Federal Register Notice summarizes 
the fastidious efforts taken to examine other potential variables. For 
each of the other variables considered, either insufficient evidence 
existed to determine if an adjustment should be made (e.g., for 
variables such as race and ethnic background, and population density) 
or the available evidence suggested that adjustment should not be made 
(i.e., gender). The findings from these efforts indicated that the 
prevalence of SED is greater in children from low socio-economic 
backgrounds than in children from middle-class or upper-class 
backgrounds. As a result, the decision was made to include percent-in-
poverty as an adjustment factor. While the data were clear about an 
overall relationship, in the absence of any national studies, the 
quantitative adjustment that should be made could not be determined 
with precision. It therefore was decided that since the report could 
offer only general estimates of prevalence, given the shortcomings of 
the available data, the simplest and perhaps clearest way to adjust for 
percent-in-poverty would be to divide the States into groups based on 
the percent-in-poverty. Although this ``grouping'' method may 
potentially exaggerate the differences between States that fall in 
different categories, the percent-in-poverty measures differ in a 
relatively minor way. Because the estimates are not to be used to 
determine funding levels, the decision was made to use this grouping 
method despite minor problems. It is hoped that additional research 
will permit more precise estimations in the future.
    With regard to estimation methods, concerns were also raised that 
the selection of poverty as the only variable to ``correct'' the 
estimated prevalence of SED would produce data that underestimated the 
State prevalence rates of SED. Several States emphasized that 
additional factors, including geographical data (urban/rural), would 
provide more representative data. CMHS recognizes the importance of 
this data. However, presently, the data in this area is not precise 
enough to draw estimates; in the absence of a national study, CMHS 
chose to utilize and analyze the most precise data available. In this 
instance, percent-in-poverty rates proved to be the most precise data 
available. As new data become available, these issues will be 
revisited.
    One comment raised specific questions about the comparability of 
the prevalence estimates for children with SED with estimates from 
other studies. For example, Knitzer, in ``At the Schoolhouse Door,'' 
estimates that 3 to 5 percent of children are ``judged to be seriously 
emotionally disturbed'' (p. xii). However, this book was published in 
1990, before CMHS developed the definition of SED on which the present 
estimate is based and before the results of most of the studies 
included in the present report were available. Similarly, the 1969 
Joint Commission on the Mental Health of Children indicates that 2 to 3 
percent suffered from severe disorders. The present report is based not 
only on more recent data but also on new instruments and a revised 
diagnostic system.
    Finally, concerns were raised that prevalence estimates for 
children/adolescent with SED in individual States are not uniformly 
consistent with estimates for adults with SMI published by CMHS. In 
comparing data for children and adults, it should be

[[Page 38663]]

remembered that the data for children cover a restricted period of nine 
years (from ages nine through 17) while the adult estimates are for the 
adult lifetime, beginning at age 18 and over. Therefore, it is not 
surprising that within the same State estimates for children may be 
lower or higher than adults. Further, the group of technical experts 
that developed estimates for SMI found substantially higher prevalence 
rates in young adults than in older adults. Consequently, States with a 
high percentage of elderly will have lower overall prevalence rates of 
SMI than will States with a high percentage of young adults. When 
comparing adult prevalence rates with those for children, it is 
important to remember that the children's data are based on a 
relatively short developmental stage in relation to the adult rates.

Exclusion of Children Age Birth to 8

    Several comments acknowledged the paucity of research on children 
from birth to 8 years and inquired about future research efforts by 
CMHS to address this population. CMHS acknowledges the need to develop 
estimation methodology for this very important population of young 
children. Current plans for developing this methodology include an 
updated literature review of prevalence data for children with a SED in 
the birth to 8 age group. CMHS will make these data available when 
obtained.

Exclusion of Puerto Rico

    It was brought to the attention of CMHS that there was significant 
interest in obtaining prevalence estimates for children with SED in 
Puerto Rico. Estimates of children with SED, published on Monday, 
October 6, 1997, in Federal Register, Notice Volume 62, No 193, p. 
52139, were based on 1995 U.S. Census Bureau population and poverty 
rate data. These Census Bureau estimates are not available for Puerto 
Rico and other U.S. territories. CMHS responds to these comments by 
obtaining SED estimates for Puerto Rico derived from 1990 census data 
(the most recent year for which data are available).
    According to the Census Bureau, the poverty rate for Puerto Rico in 
1990 was 66.8 percent for persons under 18 years. Using the steps 
outlined on page 52141 of the above Federal Register Notice, Puerto 
Rico with a poverty rate of 66.8 percent will be included in group C 
(the group with poverty rates in excess of 22 percent). At a level of 
functioning of 50 (LOF=50), the number of children and adolescents with 
SED is estimated to be between 7-9 percent of youth 9-17 years of age. 
At a level of functioning of 60 (LOF=60), the number of children and 
adolescents with SED is estimated to be between 11-13 percent of youth 
9-17 years of age.

 Table 1.--Estimates of Children and Adolescents With Serious Emotional Disturbance; State Estimates Algorithms 
----------------------------------------------------------------------------------------------------------------
                                                                    LOF*=50                     LOF*=60         
          Territory             Number of    Percent in  -------------------------------------------------------
                               youth 9-17      poverty     Lower limit   Upper limit   Lower limit   Upper limit
----------------------------------------------------------------------------------------------------------------
Puerto Rico.................      602,309          66.8        42,162        54,208        66,254       78,300  
----------------------------------------------------------------------------------------------------------------
* LOF=Level of functioning from Children's Global Assessment Scale.                                             

Exclusion of Substance Use Disorders

    The decision to exclude substance use disorders from this 
estimation methodology was addressed in the 1993 Federal Register 
Notice that provided a national definition of SED. Because substance 
use disorders are not included in the definition of serious emotional 
disorder, they are not included in the current estimation methodology. 
Please see the Federal Register Notice (1993, 58(96), p. 29424) for a 
more detailed explanation.

Instrumentation

    CMHS stresses that the methodology is based on the Children's 
Global Assessment Scale (CGAS) because the CGAS was the most commonly 
used instrument found in the community-based epidemiology literature 
received by the group of technical experts. When other instruments were 
used, the findings were taken into consideration. CMHS recognizes that 
a number of States use the Children's Adolescent Functional Assessment 
Scale-Mini-Scale and, consequently, does not discourage the use of this 
instrument.

Definition of Serious Emotional Disturbance

    Some States expressed concern that the definition of SED used to 
estimate prevalence may result in an over-estimate of prevalence by 
counting children who had a diagnosis and functional impairment over a 
2-year period rather than a 1-year period.
    The definition used to estimate prevalence is ``total number of 
cases in a year.'' None of the studies cited in the report gathered 
prevalence information of a duration of greater than a year. In fact, 
most of the studies used to formulate the prevalence estimates utilized 
the Diagnostic Interview Schedule for Children, which derives 
prevalence information for a 6-month time period. Therefore, not only 
does the definition ensure against an over estimate of prevalence but 
also there is a possibility of a slight under estimate, based on the 
methods used.

Estimation Procedures

    The following steps were taken to adjust for differences in State 
socio-economic circumstances. The 1995 State-by-State estimates of 
children and adolescents with SED are provided in Table 3.

Step 1

    States were sorted by poverty rates (1995), in ascending order. 
Using this sort order, States were initially classified into three 
groups of equal proportions, i.e., the first 17 States were put into 
Group A; the next 17 States, into Group B; the remaining 17 States, 
into Group C. However, in reviewing the results, we noted that 
observations 17 and 18 differed by .01 percent. Observation number 18 
was included in group A. For this reason, Group A has 18 cases, Group B 
has 16 cases, and Group C has 17 cases. Group A is the lowest 
percentage of children in poverty; Group B represents a mid-point; and 
Group C includes the highest percentage of children in poverty.

Step 2

    At a level of functioning of 50 (LOF=50), the number of children 
and adolescents with SED is calculated to be between 5-7 percent of the 
number of youth between 9-17 years for Group A. For Group B, the 
estimate is between 6-8 percent of the number of youth 9-17 years. The 
estimated SED population for

[[Page 38664]]

Group C is calculated to be between 7-9 percent of the number of youth 
9-17 years.

Step 3

    At a level of functioning of 60 (LOF=60), the number of children 
and adolescents with SED is calculated to be between 9-11 percent of 
the number of youth 9-17 years for Group A. For Group B, the estimate 
is between 10-12 percent of the number of youth 9-17 years. The 
estimated SED population for Group C is calculated to be between 11-13 
percent of the number of youth 9-17 years.

    Table 2.--1995 Estimates of Children and Adolescents With Serious Emotional Disturbance; State Estimates    
                                                   Algorithms                                                   
----------------------------------------------------------------------------------------------------------------
                                                                             Estimated population               
                                                             ---------------------------------------------------
                           States                                      LOF*=50                   LOF*=60        
                                                             ---------------------------------------------------
                                                              Lower limit  Upper limit  Lower limit  Upper limit
----------------------------------------------------------------------------------------------------------------
Group A Lowest percent in poverty...........................           5%           7%           9%          11%
Group B Medium percent in poverty...........................           6%           8%          10%          12%
Group C Highest percent in poverty..........................           7%           9%          11%         13% 
----------------------------------------------------------------------------------------------------------------
* LOF=Level of functioning from the Children's Global Assessment Scale.                                         


         Table 3.--1995 Estimates of Children & Adolescents With Serious Emotional Disturbance by State         
----------------------------------------------------------------------------------------------------------------
                                                                       LOF*=50                   LOF*=60        
               State                 Number of    Percent in ---------------------------------------------------
                                     youth 9-17    poverty    Lower limit  Upper limit  Lower limit  Upper limit
----------------------------------------------------------------------------------------------------------------
      Total.......................   33,706,204  ...........    2,118,269    2,792,391    3,466,516    4,140,636
 1  New Hampshire.................      147,695         4.07        7,385       10,339       13,293       16,246
 2  Alaska........................       90,955         8.96        4,548        6,367        8,186       10,005
 3  New Jersey....................      932,671         9.60       46,634       65,287       83,940      102,594
 4  Utah..........................      349,086         9.76       17,454       24,436       31,418       38,399
 5  Minnesota.....................      643,892        11.30       32,195       45,072       57,950       70,828
 6  Colorado......................      491,930        11.34       24,597       34,435       44,274       54,112
 7  Nebraska......................      231,037        11.62       11,552       16,173       20,793       25,414
 8  Missouri......................      709,439        11.74       35,472       49,661       63,850       78,038
 9  Kansas........................      354,722        12.55       17,736       24,831       31,925       39,019
10  Wisconsin.....................      706,004        12.56       35,300       49,420       63,540       77,660
11  Hawaii........................      143,901        13.97        7,195       10,073       12,951       15,829
12  North Dakota..................       91,443        14.13        4,572        6,401        8,230       10,059
13  Virginia......................      790,359        14.38       39,518       55,325       71,132       86,939
14  Nevada........................      186,695        14.41        9,335       13,069       16,803       20,536
15  Indiana.......................      758,633        15.24       37,932       53,104       68,277       83,450
16  Rhode Island..................      115,176        15.36        5,759        8,062       10,366       12,669
17  Delaware......................       85,396        15.56        4,270        5,978        7,686        9,394
18  Maine.........................      160,434        15.57        8,022       11,230       14,439       17,648
19  Vermont.......................       76,500        15.79        4,590        6,120        7,650        9,180
20  Maryland......................      608,209        15.80       36,493       48,657       60,821       72,985
21  Wyoming.......................       75,106        16.21        4,506        6,008        7,511        9,013
22  Georgia.......................      942,161        16.30       56,530       75,373       94,216      113,059
23  Massachusetts.................      680,101        17.12       40,806       54,408       68,010       81,612
24  Iowa..........................      385,583        17.39       23,135       30,847       38,558       46,270
25  Washington....................      714,567        17.81       42,874       57,165       71,457       85,748
26  Connecticut...................      378,473        18.03       22,708       30,278       37,847       45,417
27  Pennsylvania..................    1,462,731        18.07       87,764      117,018      146,273      175,528
28  Oregon........................      411,543        18.22       24,693       32,923       41,154       49,385
29  Michigan......................    1,275,452        18.36       76,527      102,036      127,545      153,054
30  Ohio..........................    1,451,220        19.33       87,073      116,098      145,122      174,146
31  Idaho.........................      183,829        20.57       11,030       14,706       18,383       22,059
32  South Dakota..................      108,855        20.74        6,531        8,708       10,886       13,063
33  North Carolina................      879,091        21.06       52,745       70,327       87,909      105,491
34  Kentucky......................      504,373        21.25       30,262       40,350       50,437       60,525
35  Illinois......................    1,517,182        22.14      106,203      136,546      166,890      197,234
36  Tennessee.....................      658,573        22.23       46,100       59,272       72,443       85,614
37  Montana.......................      126,834        22.39        8,878       11,415       13,952       16,488
38  Arkansas......................      337,718        22.44       23,640       30,395       37,149       43,903
39  Texas.........................    2,623,654        24.53      183,656      236,129      288,602      341,075
40  California....................    3,968,950        24.97      277,827      357,206      436,585      515,964
41  Oklahoma......................      457,496        24.98       32,025       41,175       50,325       59,474
42  Arizona.......................      542,019        25.31       37,941       48,782       59,622       70,462
43  Florida.......................    1,623,697        25.50      113,659      146,133      178,607      211,081
44  New York......................    2,141,435        25.51      149,900      192,729      235,558      278,387
45  West Virginia.................      231,390        26.93       16,197       20,825       25,453       30,081
46  Alabama.......................      547,671        27.50       38,337       49,290       60,244       71,197
47  Louisiana.....................      639,158        29.69       44,741       57,524       70,307       83,091

[[Page 38665]]

                                                                                                                
48  South Carolina................      470,875        32.11       32,961       42,379       51,796       61,214
49  Washington, DC................       48,365        35.33        3,386        4,353        5,320        6,287
50  New Mexico....................      251,231        36.59       17,586       22,611       27,635       32,660
51  Mississippi...................      392,694        37.03       27,489       35,342       43,196       51,050
----------------------------------------------------------------------------------------------------------------

    Dated: June 29, 1998.
Joseph Faha,
Director, Legislation & External Affairs.
[FR Doc. 98-19039 Filed 7-16-98; 8:45 am]
BILLING CODE 4160-20-U