Improved Accuracy of PSA Self-Reports Among Low-income, Uninsured Men after a Public Health Nursing Intervention

background:  We examined the accuracy of patient-reported prostate-specific antigen (PSA) levels among indigent, uninsured men in a state-funded prostate cancer treatment program providing case management, care coordination and health education. We made assessments at enrollment and again at the next PSA measurement to determine accuracy before and after exposure to the program’s nursing interventions, designed to target barriers to care including language and education.

methods:  Among enrolled men, we sampled 131 with matched self-reported and lab-reported PSA levels for the same lab draw within 3 months of program entry (baseline).  Of these, 62 had a second matched self-report and lab-report PSA measurement within 6 months of the first PSA.  Patient-reported values were compared with lab-reported results (ng/mL) as absolute differences.  Characteristics were compared between men categorized as accurate (zero difference) or inaccurate (difference >0) at each time point.   

results:  At baseline, men averaged 59.2 years of age and 50% were Hispanic. 41% (n=54) of men reported accurate values. Among the 77 (59%) with inaccurate self-reports, the mean (SD) absolute difference was 17.2 (106.2) and the median 0.90. Despite no difference in age, race, language, income, and comorbidity between both groups, language was found significant in the multivariate regression.  Men with inaccurate self-reports were more likely to be English-speaking (OR=0.38, 95% CI 0.14-0.88). At follow-up, 56% (n=34) were accurate vs. 41% at baseline (p=0.0602).

conclusions: Exposure to the program’s nursing intervention was associated with improved PSA reporting accuracy.  English speakers less accurately reported PSA results at baseline.