Why were these statistics labelled as experimental?

The NHS Information Centre first released these statistics in October 2008 as 'experimental statistics', reflecting the fact the methodologies for deriving a number of the statistics were new, and there were some definitional and data quality issues.

Although MHMDS data has been collected since April 2003 there had been little reporting using it at a national level. The resultant lack of feedback limited provider trusts' ability to improve data quality. In addition to issues of coverage and completeness the definition of the dataset is not properly representative of all mental health care activities. This caused difficulties with data collection and limited the usefulness of all the data. There were also issues with the processing of the data that had an impact on the accuracy of record and person counts.

However, with this fifth release of annual statistics from MHMDS the methodology for most of the analyses is established and the statistics presented here are no longer considered as experimental.

This release introduces an element of benchmarking in so far as some commissioner level figures can be compared with the average for ONS peer group organisations, but the main aim of MHMDS Online remains to encourage provider and commissioner organisations to review the information about their own organisation.

Year on year comparisons
Only the data that are considered more reliable are used in these statistics. However, any year on year comparisons should be treated with great caution as some changes will be due to data quality issues and the data for 3 years 2003-06 were processed and validated by a different method than that currently being used by the NHS Information Centre.

The NHS IC now publishes trust level reports on MHMDS data quality for each quarter and for the annual MHMDS submissions and these can be downloaded from: www.ic.nhs.uk/services/mhmds/dq. Further information about data quality can also be found in the supporting document MHMDS Experimental Statistics: Data Quality and Methodology.

The organisation continues to work with trusts and other interested parties to improve the quality and coverage of the data and to address any issues as they arise. Please feel free to contribute to this work by emailing your comments to The NHS Information Centre's Mental Health Team at mhmds@ic.nhs.uk.

Close this notice panel