Table 1 The German Network Health Services Research (www.dnvf.de) criteria for assessment of registry quality. For details see [17]

1. Systematics and Appropriateness

Registry protocol

Aims and expected benefits

Registry organization

Patient rights and data safety

Registry design

Data acquisition and processing

Data analyses, reporting and publication

Steering Committee

2. Standardization

Definitions and standard operating procedures

Training for data entry and verification

3. Validity of the sampling procedure

Definition of inclusion and exclusion criteria

Completeness and representativity of data

Consideration of sample size and effect

Consideration of confounders and biases

4. Validity of data collection

Assessment of data: completeness, plausibility, distribution, Concordance

Handling of missing data, follow-up and dropout

Unique identifiers and risk of duplicate data entry

Monitoring and audits

5. Validity of statistical analyses and reports

Patient/data flow scheme

Handling of missing data

Assessment of baseline data and outcome measures, Evaluation of balance in comparative analyses

Assessment of precision measures

Methods against bias and confounders

Adjustment for multiple inference testing

Assessment of relative and absolute effects, adjusted and Unadjusted results

Multivariate modeling for complex questions

Consideration of timeline in longitudinal data

Control of cluster effects

Validity of statistical analyses and reports

6. General demands for registry quality

Transparency towards limitations

Acceptance among patients and institutions

Transparency and scientific independence

Flexibility and adaptability

Topicality