Validation of data quality in the Swedish National Register for Oesophageal and Gastric Cancer. Issue 10 (28th July 2016)

• Record Type: Journal Article
• Authors: Linder, G.; Lindblad, M.; Djerf, P.; Elbe, P.; Johansson, J.; Lundell, L.; Hedberg, J.
• Is Part Of: British journal of surgery. Volume 103:Issue 10(2016)
• Page Start: 1326
• Page End: 1335
• Publication Date: 2016-07-28
• Subjects: Surgery -- Periodicals; 617.005
• DOI: 10.1002/bjs.10234
• Languages: English
• ISSNs: 0007-1323

Abstract

Abstract : Background: The Swedish National Register for Oesophageal and Gastric Cancer (NREV) was launched in 2006. Data are reported at diagnosis (diagnostic survey), at the time of surgery (surgical survey) and at first outpatient follow‐up (follow‐up survey). The aim of this study was to evaluate data originating from NREV in terms of comparability, completeness, accuracy and timeliness. Methods: Coding routines were compared with international standards and completeness was evaluated by means of a 5‐year (2009–2013) comparison with mandatory national registers. Validity was tested by comparison with reabstracted data from source medical records in 400 patients chosen randomly with stratification for hospital size and catchment area population. Timeliness of registration was described. Results: Coding routines followed national and international guidelines. Compared with the Swedish Cancer Registry from 2009 to 2013, 6069 (95·5 per cent) of 6354 patients were registered in NREV at the time of data extraction. Of 60 variables investigated, 10 966 of 12 035 original entries were correct in the reabstraction, resulting in an exact agreement of 91·1 per cent in the register. There were 782 (6·5 per cent) incorrect and 287 (2·4 per cent) missing entries. Median time to registration was 3·9, 3·4 and 4·1 months for diagnostic, surgical and follow‐up surveys respectively. Conclusion: NREV has reached a position with good coverage of those with the relevant diagnoses, and contains comparable and valid data. Quality data on each variable are available. Timeliness is an area with potential for improvement. Abstract : National databases can be highly accurate