Part III
Prevalence of Selected Mental Disorders
The task order requires the committee to compare trends in the prevalence of mental disorders in children in the United States with trends in the Supplemental Security Income (SSI) childhood disability population. Part III of the report responds to the task order by providing information on the population-based prevalence estimates and prevalence trends of selected childhood mental disorders which are the most frequently reported cause of disability in the SSI program. The data presented here represent a selection of the data that are directly relevant to the fundamental issue that confronts the SSI disability benefit program for children: whether or not the trends that are being observed in the SSI program are out of line with trends observed in the general population. This introduction provides background information relevant for interpretation of the chapters included in Part III of this report.
COMPARING PREVALENCE TREND DATA WITH SSI DATA
As described in Chapter 2, the trends in the SSI program are not directly comparable to prevalence trends otherwise generated from the general population. Comparisons of SSI trends with any of the data included in the following chapters should take into consideration the major differences outlined in Chapter 2, including the differences between the composition and characteristics of the SSI population versus the general population and the methodological differences in ascertainment and classification of cases of mental disorders in children.
The availability and quality of prevalence trend data for mental disorders
in children vary depending on the specific mental disorder. Unfortunately, for many conditions neither cross-sectional prevalence data nor prevalence trend data are available. Each of the following chapters will reference some of the best available population-based data that can be compared to trends in the SSI program for children with mental disorders. As briefly explained in Chapter 2, the sources of relevant data in this report include administrative data, surveillance data, national surveys involving direct assessment or parent report, published meta-analyses, and Medicaid data. Also as explained in Chapter 2, the characteristics of the prevalence data used in this report vary depending on their source.
Generally, prevalence trend data for pediatric mental disorders, based on direct neuropsychological assessment or other systematic measurement of a child’s symptoms, is unavailable. In addition, data on patterns and trends of the severity of mental disorders in children are unavailable. With a few exceptions the data referenced in Part III of the report rely on proxies for determining whether a child has a mental disorder, such as parent report of whether a not a child has received diagnosis, whether a child is receiving treatment, or whether the child is receiving special education services for a disorder. As discussed in more detail in Chapter 2, because these methods for identifying cases of a disorder do not directly assess the symptoms of a child, they can be susceptible to error or bias, depending on how cases are ascertained. For example, data based on the receipt of treatment can be biased, because many children who receive treatment may not meet the diagnostic criteria for having a disorder, or conversely many children who have the disorder may not receive treatment. As a result, use of these data sources as a reference to the frequency of a disorder within the population has some limitations. These limitations are discussed in Chapter 2. Additional information about, and references for the methods used in, the national surveys can be found in Appendix D.
In spite of these limitations, the committee concluded the data reviewed in Part III are necessary to meet the requirements of the task order for several reasons. First, the information reviewed by disability examiners during the adjudication process is similar to the information referenced in this report. In particular, the use of special education services and parent reports of diagnoses and symptoms are the same types of data frequently referenced by disability examiners during the adjudication process. Second, disability examiners are not clinicians or researchers, and do not directly assess the symptoms of applicants. Rather, disability examiners are the Social Security Administration (SSA) staff trained to apply regulatory standards to evidence provided by applicants and other medical experts who may directly examine the applicant. The disability adjudication process more closely resembles survey methodology than a clinical assessment. Finally,
since no other data are available, every resource should be included to provide the best available information for the SSA’s use.
COMORBIDITY DATA
As discussed in Chapter 2, the SSI data contain no usable comorbidity data. As a result, patterns of comorbidity within the SSI population cannot be reliably assessed and cannot be compared with the general population. Descriptions of comorbidity patterns for individual mental disorders in children can be found in Part II of the report. An analysis of the comorbidity within the SSI-eligible population of child Medicaid enrollees with mental disorders is included in Part IV of the report.
RACE AND ETHNICITY DATA
As discussed in Chapter 2, there are no data available from SSA about the race and ethnicity of child SSI recipients. As result, trends in the rates of mental disorders, by race and ethnicity, are not discussed in Part III of the report. Data on the rates of mental disorder diagnoses, by race and ethnicity, among children with disabilities enrolled in Medicaid in 2010, are briefly discussed in Part IV of the report.
MEDICAID DATA
Special notice should be paid to the Medicaid data included in this report. Appendixes F and G contain a report and analysis commissioned by the Institute of Medicine committee and performed by the Center for Health Services Research on Pharmacotherapy, Chronic Disease Management, and Outcomes and the Center for Education and Research on Mental Health Therapeutics at the Rutgers University Institute of Health. The Medicaid report contains an analysis of Medicaid eligibility, service use, and payment data that provides novel insight into the rates of mental disorder diagnoses and treatment among all children who are enrolled in Medicaid as well as among the children who are enrolled in Medicaid on the basis of qualifying for and receiving SSI disability benefits. Each of the following chapters in Part III of the report will reference findings contained in the Medicaid report.
As described in Chapter 2, the Medicaid Analytic eXtract (MAX) provides estimates for national trends in the diagnosis and treatment of mental disorders for children and adolescents living in low-income households. The impetus for the Medicaid analysis came from committee deliberations, where committee members identified several challenges posed by the task order. One issue, as previously articulated in Chapters 2 and 5, is that the
vast majority of the SSI population consists of children in families who are near or below the federal poverty level, while the general population is not. The committee determined that MAX data could potentially provide a unique perspective because Medicaid is a means-tested program and, as such, the socioeconomic status of the population of children who are enrolled in Medicaid will be more directly comparable to the socioeconomic status of the population of children who are on SSI.
Each of the following chapters in Part III includes four sections:
- A review of estimates of prevalence of the mental disorder from the general population.
- An analysis of trends from the Medicaid report.
- A summary of trends of allowances and recipients for the mental disorder in question from 2004 to 2013.
- Conclusions and findings, including a rough estimate of the proportion of potentially eligible children and those who are recipients of SSI benefits.