Email updates

Keep up to date with the latest news and content from BMC Research Notes and BioMed Central.

Open Access Research article

Valid comparisons and decisions based on clinical registers and population based cohort studies: assessing the accuracy, completeness and epidemiological relevance of a breast cancer query database

Christian Olaf Jacke1*, Mathias Kalder2, Uwe Wagner2 and Ute-Susann Albert2

Author Affiliations

1 Mental Health Services Research Group, Central Institute of Mental Health, Medical Faculty Mannheim / Heidelberg University, Mannheim, Germany

2 Department of Gynecology, Gynecological Endocrinology and Oncology, Breast Center Regio, University of Marburg, Marburg, Germany

For all author emails, please log on.

BMC Research Notes 2012, 5:700  doi:10.1186/1756-0500-5-700

Published: 27 December 2012

Abstract

Background

Data accuracy and completeness are crucial for ensuring both the correctness and epidemiological relevance of a given data set. In this study we evaluated a clinical register in the administrative district of Marburg-Biedenkopf, Germany, for these criteria.

Methods

The register contained data gathered from a comprehensive integrated breast-cancer network from three hospitals that treated all included incident cases of malignant breast cancer in two distinct time periods from 1996–97 (N=389) and 2003–04 (N=488). To assess the accuracy of this data, we compared distributions of risk, prognostic, and predictive factors with distributions from established secondary databases to detect any deviations from these “true” population parameters. To evaluate data completeness, we calculated epidemiological standard measures as well as incidence-mortality-ratios (IMRs).

Results

In total, 12% (13 of 109) of the variables exhibited inaccuracies: 9% (5 out of 56) in 1996–97 and 15% (8 out of 53) in 2003–04. In contrast to raw, unstandardized incidence rates, (in-) directly age-standardized incidence rates showed no systematic deviations. Our final completeness estimates were IMR=36% (1996–97) and IMR=43% (2003–04).

Conclusion

Overall, the register contained accurate, complete, and correct data. Regional differences accounted for detected inaccuracies. Demographic shifts occurred. Age-standardized measures indicate an acceptable degree of completeness. The IMR method of measuring completeness was inappropriate for incidence-based data registers. For the rising number of population-based health-care networks, further methodological advancements are necessary. Correct and epidemiologically relevant data are crucial for clinical and health-policy decision-making.

Keywords:
Data quality; Data quality indicators (DQI); Data correctness; Data accuracy; Data completeness