Abstract
A98
African Americans continue to experience higher cancer incidence (all sites combined), higher mortality rates, lower rates of cancer survival, and are less likely to participate in cancer screening compared to other racial / ethnic groups in the United States. The medical record is viewed as the legal standard by which care is documented and evaluated. Self-reports are often debated in terms of its reliability and accuracy. Research that compared self-reported screening to the medical record documentation found that participants overestimate their participation in cancer screening and are inaccurate when recalling the timeframe for past screenings. This descriptive study, guided by the Patient / Provider / System Theoretical Model (PPSTM) for cancer screening, evaluates levels of incongruence (cases where persons self reported screening differed from the medical record), specificity (cases where persons reported not being screened and the medical record documents no screening), and sensitivity (cases where persons reported being screened and the medical record documented screening) between self-reported screening and medical record documentation for breast, cervical, and colorectal cancer screening. Data were collected from African American women (N=116) interviewed at federally qualified health centers (FQHC) using the Patient Demographic Data Questionnaire and a Medical Record Abstraction Form. Women were eligible to participate in the study if they were 18 years and able to understand English or Spanish. They were recruited from the waiting areas of the FQHC’s over a consecutive 5-day period. Medical records were reviewed by registered nurses. The mean age was 35 years and their mean educational level was 12.6 years. The majority (61%) reported annual incomes of less than $10,000. The overall level of incongruence ranged from 6% to 59% across the screening types and sensitivity ranged from 0-33%. These finding suggest that women tended to over-report screening for those procedures within the past year when compared to the medical record. There were high levels of specificity suggesting that both the participant and the medical record did not report screening, particularly during the timeframe of 2-5 years and greater than 5 years. With the exception of pap testing (timeframe of more than 5 years ago), the majority of incongruence (6-52%) is accounted for by participants who report screenings that are not documented in the medical record. More research is needed to better understand the incongruence betweenthe self-report and the medical record documentation. Several national surveys (such as the Behavioral Risk Factor Surveillance System and the National Health and Nutrition Examination Survey) depend on self-reported screening information which may influence public policy. While it is not possible to perform medical record reviews on a national level, it may be possible to develop more efficient ways to help participants track and report health care procedures. Research should also target and evaluate reasons for the high levels of specificity that suggest routine cancer screenings were not being done. These inquiries may be even more significant for African Americans who bare a disproportionate cancer burden.
First AACR International Conference on the Science of Cancer Health Disparities-- Nov 27-30, 2007; Atlanta, GA