Colonoscopy Findings in HIV-Infected Men and Women from an Urban U.S. Cohort Compared with Non-HIV-Infected Men and Women.

INTRODUCTION: As HIV-infected patients live longer, non-AIDS-defining cancers are now a major cause of morbidity and mortality. The purpose of this study was to compare the prevalence, type, and location of colorectal neoplastic lesions found on colonoscopy in HIV-infected patients from an urban U.S. cohort with non-HIV-infected patients. METHODS: We collected clinical data and colonoscopy findings on 263 HIV-infected patients matched with 657 non-HIV-infected patients on age, race, and sex. Frequency distributions and descriptive statistics were used to characterize the study population. The primary exposure was HIV infection, and the primary outcome was any adenoma or adenocarcinoma. Logistic regression models were used to estimate odds ratios with 95% confidence intervals (CIs). RESULTS: Participants were primarily African American and 40% were women. HIV-infected patients were less likely to have any neoplastic lesions (21.3% vs. 27.7%, p < .05), adenoma (20.5% vs. 27.1%, p = .04), tubular adenomas >10 mm (0.4% vs. 2.9%, p = .02), and serrated adenomas (0.0% vs.2.6%, p = <.01). There was a nonsignificant increased prevalence of adenocarcinoma in HIV-infected individuals compared with non-HIV-infected individuals (1.5% vs. 0.8%, p = .29). The lower prevalence of any adenoma remained after controlling for age, sex, smoking status, body-mass index, and diabetes mellitus [adjusted odds ratio (aOR), 0.61; 95% CI, 0.43-0.88]. HIV-infected patients had a lower prevalence of colorectal neoplastic lesions, including high-risk adenomas, than non-HIV-infected patients. CONCLUSIONS: Our findings suggest that HIV infection in a primarily African American population is associated with a lower prevalence of colorectal adenomas, but not adenocarcinoma, found by colonoscopy.

Cancer Knowledge and Opportunities for Education Among HIV-Infected Patients in an Urban Academic Medical Center.

HIV-infected patients frequently present with advanced stage cancer. It is possible that late stage presentation may be related to lack of cancer knowledge and/or barriers to care. Questionnaires were administered to 285 adult HIV-infected patients to evaluate knowledge of cancer risk factors and symptoms and barriers to care between 2011 and 2012. Differences in mean and percent scores by group were assessed using a t test for independent samples and chi-square analysis, respectively. Respondents were predominantly male (64%), African-American (86%), and low income (60% < $10,000/year). Thirty-four (12%) had been diagnosed with cancer, and 169 (59%) had a family history of cancer. The mean knowledge score was 17.5 out of 24 questions (73%). Mean scores were not significantly different by sex, age, race, or income. Respondents with a college education scored significantly higher than those with less than a high school education (p < 0.01). In unadjusted analysis, a higher proportion of patients with a personal/family history of cancer (74%) scored in the highest quartile (>70% correct) compared to those without any personal history of cancer (62%) (p = 0.03). There was a higher level of cancer knowledge in this population compared to studies that have evaluated the HIV-uninfected population. Nevertheless, there were knowledge deficits, suggesting the need for further education about cancer to improve earlier detection rates and, ultimately, outcomes.