Background: In clinical research information can be retrieved through various sources. The aim is to evaluate the agreement between answers in patient questionnaires and clinical reports in a study of patients after radical prostatectomy and patient characteristics associated with agreement between these two data sources. Methods: In the prospective non-randomized longitudinal trial LAParoscopic Prostatectomy Robot Open (LAPPRO) 4003 patients undergoing radical prostatectomy at 14 centers in Sweden were followed. Analysis of agreement is made using a variety of methods, including the recently proposed Gwet's AC1, which enables us to handle the limitations of Cohen's Kappa where agreement depends on the underlying prevalence. Results: The incidence of postoperative events was consistently reported higher by the patient compared with the clinical reports for all outcomes. Agreement regarding the absence of events (negative agreement) was consistently higher than agreement regarding events (positive agreement) for all outcome variables. Overall impression of agreement depends on which measure used for the assessment. The previously reported desirable properties of Gwet's AC1 as well as the patient characteristics associated with agreement were confirmed. Conclusion: The differences in incidence and agreement across the different variables and time points highlight the importance of carefully assessing which source of information to use in clinical research. Trial registration: ISRCTN06393679 (www.isrctn.com). Date of registration: 07/02/2008. Retrospectively registered.
Subject classification (UKÄ)
- Urology and Nephrology
- Public Health, Global Health, Social Medicine and Epidemiology
- Case-report form
- Prostate cancer