OBJECTIVES: The objective of the study was to compare the effect of two approaches to requesting medical records on survey response rates, sample representativeness, and the quality of self-reported screening. STUDY DESIGN AND SETTING:Eight hundred ninety veterans aged 50-75 years from the Minneapolis VA Medical Center were randomly assigned to (1) records request included with a colorectal cancer screening survey ("with-survey" group) or (2) request in a separate mailing following a completed survey ("after-survey" group). Analyses compared response rates, the proportion and characteristics of patients providing records, and the validity of self-reported screening, by group. RESULTS:Response rates did not vary by group (with-survey 76%; after-survey 78%, P=0.45). 54% of with-survey and 47% of after-survey participants provided complete medical records (P = 0.06). In the with-survey group, patients with complete medical records were significantly more likely to be married and to have a diagnosis of posttraumatic stress disorder; in the after-survey group, they were more likely to be aged 65-75 years, Caucasian, to have a family history of colorectal cancer, and to report being screened. Validity of self-reported screening did not vary significantly by group. CONCLUSION: The with-survey approach did not significantly reduce response rates or the quality of self-reported screening and produced a higher number and more representative sample with complete records.
RCT Entities:
OBJECTIVES: The objective of the study was to compare the effect of two approaches to requesting medical records on survey response rates, sample representativeness, and the quality of self-reported screening. STUDY DESIGN AND SETTING: Eight hundred ninety veterans aged 50-75 years from the Minneapolis VA Medical Center were randomly assigned to (1) records request included with a colorectal cancer screening survey ("with-survey" group) or (2) request in a separate mailing following a completed survey ("after-survey" group). Analyses compared response rates, the proportion and characteristics of patients providing records, and the validity of self-reported screening, by group. RESULTS: Response rates did not vary by group (with-survey 76%; after-survey 78%, P=0.45). 54% of with-survey and 47% of after-survey participants provided complete medical records (P = 0.06). In the with-survey group, patients with complete medical records were significantly more likely to be married and to have a diagnosis of posttraumatic stress disorder; in the after-survey group, they were more likely to be aged 65-75 years, Caucasian, to have a family history of colorectal cancer, and to report being screened. Validity of self-reported screening did not vary significantly by group. CONCLUSION: The with-survey approach did not significantly reduce response rates or the quality of self-reported screening and produced a higher number and more representative sample with complete records.
Authors: Timothy J Beebe; Jeanette Y Ziegenfuss; Sarah M Jenkins; Lindsey R Haas; Michael E Davern Journal: Ann Epidemiol Date: 2011-04-16 Impact factor: 3.797
Authors: S Trent Rosenbloom; Jennifer L Madison; Kyle B Brothers; Erica A Bowton; Ellen Wright Clayton; Bradley A Malin; Dan M Roden; Jill Pulley Journal: J Am Med Inform Assoc Date: 2013-07-25 Impact factor: 4.497
Authors: Timothy J Beebe; Jeanette Y Ziegenfuss; Jennifer L St Sauver; Sarah M Jenkins; Lindsey Haas; Michael E Davern; Nicholas J Talley Journal: Med Care Date: 2011-04 Impact factor: 2.983