Name: The Socioeconomic Status Instrument (SESI)
Abstract
Background: Because of the lack of consistency in the associations of the socioeconomic status (SES) of prostate cancer (PC) patients from diverse racial and ethnic backgrounds with PC health outcomes, Dr. Cyrus-David created the Socioeconomic Status Instrument (SESI) from the Demographic and Health Access components of the Behavioral Risk Factor Surveillance System 2004 Questionnaires and the socioeconomic indices of the subjects’ residential counties to better assess the SES of PC patients.
Methods: The SESI was tested on 220 consecutive subjects with pathologically confirmed PC at the Veterans Affairs Medical Center in Houston, TX. A team that included an epidemiologist, a validation statistician/ health services research scientist, and PC survivors assessed the content validity of the SESI. The construct validity of the SESI was assessed with factor analysis by extracting and analyzing five principal components based on the subjects’ individual responses on the assessment: county socioeconomic characteristics, individual socioeconomic characteristics, financial distress, increased domestic burden with limited earnings, and affluence. The internal consistency reliability of the SESI was assessed with Cronbach alpha coefficients. Results: Based on the reviews of the SESI, all of the initial 10 items were retained. The correlations between individual responses on the SESI were similar to the results of previous studies. The five principal components that Dr. Cyrus-David assessed accounted for 71.5% of the variance. Factor loadings ranged from 0.66 to 0.98 and communalities ranged from 0.55 to 0.94. County socioeconomic characteristics accounted for 22.6% of the variance, whereas individual socioeconomic characteristics accounted for 14.6% of the variance. The overall Cronbach alpha coefficient was 0.78.
Conclusions: The SESI is valid and reliable. Accurate measurements of the SES of PC patients would provide better guidance for future research and care deliveries. Cancer Epidemiology 34:382–387, 2010. DOI 2010.10.1016/j.canep.2010.04.020
