Mental Health Client-Level Data (MH-CLD)
Client-Level Mental Health Data
MH-CLD and the Mental Health Treatment Episode Data Set (MH-TEDS) provide information on mental health diagnoses and the mental health treatment services, outcomes, and demographic and substance use characteristics of people in mental health treatment facilities. This information comes from facilities that report to individual state administrative data systems.
To find out more or to read publications using MH-CLD information, please visit the SAMHSA Data Website.
Scope and Methodology Notes
Both MH-TEDS and MH-CLD include people served through state mental health authorities (SMHAs) during the state-defined 12-month reporting period. Most states elect to use the state fiscal year, which usually runs from July 1 through June 30 of the following year; however, some states define their fiscal year differently.
MH-CLD contains one record for each person served. Data items are populated by the states based on the person’s status, either at the start and end of the state’s reporting period or according to the most recently available data.
MH-TEDS contains one record for each admission to, and discharge from, a service type or setting (referred to as a treatment episode) within the client’s treatment history during the reporting period. Data collected from MH-TEDS admission records and update or discharge records are stored as two data sets. These data sets are linked using a HIPAA-compliant, nonprotected health information, unique client identifier and other key fields located in both files. Linking the records allows for the creation of a single client record comparable to the MH-CLD format.
MH-TEDS data is made compatible for analysis with MH-CLD data by developing a file with a structure similar to the MH-CLD data set. After transposing admission and discharge records from MH-TEDS into a client-level file, the MH-TEDS variables are crosswalked to MH-CLD variables. The process of creating an individual record using the MH-TEDS data allows for reporting of the individuals in both the MH-TEDS and MH-CLD data sets.
MH-CLD, while providing valuable information about mental health service use, does not represent the total national demand for mental health treatment or describe the mental health status of the national population. MH-CLD is a compilation of client-level data from facilities that SMHAs operate.
Different factors affect which types of facilities provide data, including differences in state licensure, certification, accreditation, and disbursement of public funds. For example, some SMHAs regulate private facilities and individual practitioners, while others do not. Moreover, in some states, mental health services are provided in correctional facilities under the guidance of mental health agencies, while in other states they are not.
Up to three mental health diagnoses per individual can be reported to MH-TEDS or MH-CLD. These diagnoses may not reflect all diagnoses for people served. Furthermore, some individuals have no valid mental health diagnosis reported. If the missing diagnosis is not randomly distributed across facilities, estimated prevalence rates of mental health diseases may be biased.
For methodological information for a particular year or date range, please check the codebook for a data set from the sidebar to the right.
The data limitations include the following:
- MH-CLD does not represent the total national demand for mental health treatment or describe the mental health status of the national population.
- The scope of facilities reporting data is limited and varies based on the state.
- The mental health diagnoses in the data set may not represent all diagnoses for individuals served. Furthermore, some individuals have no valid mental health diagnosis reported. If the missing diagnosis is not randomly distributed across facilities, estimated prevalence rates of mental health diseases may be biased.
Several measures are taken to protect the confidentiality of all records. Variables that potentially identify an individual in their raw form undergo top- or bottom-coding. For example, age as a continuous variable has the potential to identify both the youngest and oldest participants in a public release file. For this reason, age is recoded for the public use file to reduce disclosure risk.
Disclosure analysis is used to identify records that remained unique after routine recoding measures were taken to protect confidentiality. Disclosure analysis is used to detect combinations of indirect identifiers that potentially link an individual to a record. Particular attention was given to the analytic importance of geographic data and of subgroup populations. Subsequently, data swapping was applied to produce the public use file in order to satisfy strict confidentiality standards while preserving its analytic value.
The original location of a record in MH-CLD cannot be known for certain due to the use of data swapping. This method has several benefits over other disclosure protection options:
- The overall impact to the data is typically small.
- Nearly all the data are left intact.
- Data for special populations (for example, minorities) are no more impacted than other data.
- The procedures typically do not affect any analytic uses of the file.
- The procedures allow greater detail to remain on the public use file (for example, the original ethnicity codes).