DoCDat covers databases that have information on the actual or potential recipients of health care. It does not cover databases limited to the provision of resources or services (e.g. a register of hospital bed provision), useful though they are.

DoCDat focuses primarily on centralised individual-level databases, based either on prospectively or retrospectively collected data (i.e. they contain information about individuals, and centrally collate data collected from different health care providers). They must meet the following criteria:

1. Inclusion in the database is defined by a common circumstance e.g. individual's condition, intervention required or undergone (which might be a diagnostic test, treatment or a collection of interventions e.g. intensive care), administrative arrangement (such as subscriber to health insurance, registered with GP, target for immunisation), or adverse outcome (e.g. maternal death). Note that this excludes cohorts defined by behaviour or environmental exposure (for aetiological research).
   
2. Provides individual-level data, whether or not users of the database are permitted to know the identity of the individuals
   
3. Includes data from more than one provider of health care (usually many providers in a region or country). Single-provider databases may be included if they cover an area of health care for which no multi-provider database exists.
   
   

In addition, DoCDat also includes limited information on two other types of databases:

• Non-centralised individual-level databases in which the same dataset is collected by more than one provider but the data are not collated centrally.
• Aggregated databases that collect data about groups of people rather than about individuals. For example, data about the number of patients undergoing a particular operation by their age and sex, or the number of people covered in a vaccination programme.

Limited information is also provided on databases which we have not yet had a chance to assess fully. This includes the contact details for the database custodian, so that you can investigate the database further yourself.

Independent, trained interviewers assess the quality of each database using a structured questionnaire developed by clinicians, epidemiologists, statisticians and information specialists. The assessment covers:

• General aspects of the database such as when it was set up, who it includes, and what geographical area it covers.
• Data set such as how many individuals are included, data linkage, data security, patient confidentiality, and a copy of the data collection questionnaire.
• Outputs including who can analyse the data, how frequently standard audit reports are produced, and a bibliography of published work.
• Management of the database such as who is involved in running it and who funds it.
• Quality of the data including four aspects of the coverage of the data (allowing you to assess the generalisability of the data), and six aspects exploring the accuracy of the data allowing you to see how valid and reliable the data are.
• Contact details of who to contact should you wish to find out more about the database, or request access to use the data.

The definitions used in the assessment can be found in the Supporting Documentation section of these pages.