Alfred Keter

Hypertension and Obesity as Cardiovascular Risk Factors among HIV Seropositive Patients in Western Kenya

Background There is increased risk of cardiovascular disease among HIV seropositive individuals. The prevalence of HIV is highest in sub-Saharan Africa; however, HIV-related cardiovascular risk research is largely derived from developed country settings. Herein, we describe the prevalence of hypertension and obesity in a large HIV treatment program in Kenya. Methods We performed a retrospective analysis of the electronic medical records of a large HIV treatment program in Western Kenya between 2006 and 2009. We calculated the prevalence of hypertension and obesity among HIV+ adults as well as utilized multiple logistic regression analyses to examine the relationship between clinical characteristics, HIV-related characteristics, and hypertension. Results Our final sample size was 12,194. The median systolic/diastolic blood pressures were similar for both sexes (male: 110/70 mmHg, female: 110/70 mmHg). The prevalence of hypertension among men and women were 11.2% and 7.4%, respectively. Eleven percent of men and 22.6% of women were overweight/obese (body mass index ≥25 kg/m2). Ordinal logistic regression analyses showed that overweight/obesity was more strongly associated with hypertension among HIV+ men (OR 2.41, 95% CI 1.88–3.09) than a higher successive age category (OR 1.62, 95% CI 1.40–1.87 comparing 16–35, 36–45 and >45 years categories). Among women, higher age category and overweight/obesity were most strongly associated with hypertension (age category: OR 2.21, 95% CI 1.95–2.50, overweight/obesity: OR 1.80, 95% CI 1.50–2.16). Length of time on protease inhibitors was not found to be related to hypertension for men (OR 1.62, 95% CI 0.42–6.20) or women (OR 1.17, 95% CI 0.37–2.65) after adjustment for CD4 count, age and BMI. Conclusion In Western Kenya, there is a high prevalence of hypertension and overweight/obesity among HIV+ patients with differences observed between men and women. The care of HIV+ patients in sub-Saharan Africa should also include both identification and management of associated cardiovascular risk factors.

A clinician-nurse model to reduce early mortality and increase clinic retention among high-risk HIV-infected patients initiating combination antiretroviral treatment

BackgroundIn resource-poor settings, mortality is at its highest during the first 3 months after combination antiretroviral treatment (cART) initiation. A clear predictor of mortality during this period is having a low CD4 count at the time of treatment initiation. The objective of this study was to evaluate the effect on survival and clinic retention of a nurse-based rapid assessment clinic for high-risk individuals initiating cART in a resource-constrained setting.MethodsThe USAID-AMPATH Partnership has enrolled more than 140,000 patients at 25 clinics throughout western Kenya. High Risk Express Care (HREC) provides weekly or bi-weekly rapid contacts with nurses for individuals initiating cART with CD4 counts of ≤100 cells/mm3. All HIV-infected individuals aged 14 years or older initiating cART with CD4 counts of ≤100 cells/mm3 were eligible for enrolment into HREC and for analysis. Adjusted hazard ratios (AHRs) control for potential confounding using propensity score methods.ResultsBetween March 2007 and March 2009, 4,958 patients initiated cART with CD4 counts of ≤100 cells/mm3. After adjusting for age, sex, CD4 count, use of cotrimoxazole, treatment for tuberculosis, travel time to clinic and type of clinic, individuals in HREC had reduced mortality (AHR: 0.59; 95% confidence interval: 0.45-0.77), and reduced loss to follow up (AHR: 0.62; 95% CI: 0.55-0.70) compared with individuals in routine care. Overall, patients in HREC were much more likely to be alive and in care after a median of nearly 11 months of follow up (AHR: 0.62; 95% CI: 0.57-0.67).ConclusionsFrequent monitoring by dedicated nurses in the early months of cART can significantly reduce mortality and loss to follow up among high-risk patients initiating treatment in resource-constrained settings.

HIV Prevalence and Antenatal Care Attendance among Pregnant Women in a Large Home-Based HIV Counseling and Testing Program in Western Kenya

Objective To describe the uptake of and factors associated with HIV prevalence among pregnant women in a large-scale home-based HIV counseling and testing (HBCT) program in western Kenya. Methods In 2007, the Academic Model Providing Access to Healthcare Program (AMPATH) initiated HBCT to all individuals aged ≥13 years and high-risk children <13 years. Included in this analysis were females aged 13–50 years, from 6 catchment areas (11/08-01/12). We used descriptive statistics and logistic regression to describe factors associated with HIV prevalence. Results There were 119,678 women eligible for analysis; median age 25 (interquartile range, IQR: 18–34) years. Of these, 7,396 (6.2%) were pregnant at the time of HBCT; 4,599 (62%) had ever previously tested for HIV and 2,995 (40.5%) had not yet attended ANC for their current pregnancy. Testing uptake among pregnant women was high (97%). HBCT newly identified 241 (3.3%) pregnant HIV-positive women and overall HIV prevalence among all pregnant women was 6.9%. HIV prevalence among those who had attended ANC in this pregnancy was 5.4% compared to 9.0% among those who had not. Pregnant women were more likely to newly test HIV-positive in HBCT if they had not attended ANC in the current pregnancy (AOR: 6.85, 95% CI: 4.49–10.44). Conclusions Pregnant women who had never attended ANC were about 6 times more likely to newly test HIV-positive compared to those who had attended ANC, suggesting that the cascade of services for prevention of mother-to-child HIV transmission should optimally begin at the home and village level if elimination of perinatal HIV transmission is to be achieved.

Factors Associated with Uptake of Visual Inspection with Acetic Acid (VIA) for Cervical Cancer Screening in Western Kenya

Purpose Cervical cancer screening has been successful in reducing the rates of cervical cancer in developed countries, but this disease remains the leading cause of cancer deaths among women in sub-Saharan Africa. We sought to understand factors associated with limited uptake of screening services in our cervical cancer-screening program in Western Kenya. Participants and Methods Using items from a previously validated cancer awareness questionnaire repurposed for use in cervical cancer and culturally adapted for use in Kenya, we interviewed 2,505 women aged 18–55 years receiving care in gynecology clinics or seeking other services in 4 health facilities in Western Kenya between April 2014 and September 2014. We used logistic regression modeling to assess factors associated with uptake (or non-uptake), associated odds ratios (ORs) and the 95% confidence intervals (95% CI). Results Only two hundred and seventy-three women out of 2505 (11%) accepted VIA cervical cancer screening. Knowledge of just how women are screened for cervical cancer was significantly associated with reduced uptake of cervical cancer screening (OR: 0.53; CI 0.38–0.73) as was fear that screening would reveal a cancer (OR 0.70; CI 0.63–0.77), and reliance on prayer with the onset of illness (OR 0.43; CI 0.26–0.71). Participants who thought that one should get cervical cancer screening even if there were no symptoms were more than twice as likely to accept cervical cancer screening (OR 2.21; 95% CI 1.24–3.93). Older patients, patients living with HIV and women who do not know if bleeding immediately after sex might be a sign of cervical cancer were also more likely to accept screening (OR 1.03, CI 1.02–1.04; OR 1.78, CI 1.01–3.14; OR 2.39, CI 1.31–4.39, respectively). Conclusions In our population, a high percent of women knew that it is appropriate for all women to get cervical cancer screening, but only a small proportion of women actually got screening. There may be an opportunity to design educational materials for this population that will not only encourage participation in cervical cancer screening but also remediate misconceptions. The discussion illustrates how our findings could be used in such an effort.

Rates and Covariates of Recent Sexual and Physical Violence Against HIV-Infected Outpatient Drinkers in Western Kenya

Victimization from physical and sexual violence presents global health challenges. Partner violence is higher in Kenya than Africa. Violence against drinkers and HIV-infected individuals is typically elevated, so dual vulnerabilities may further augment risk. Understanding violence risks can improve interventions. Participants were 614 HIV-infected outpatient drinkers in western Kenya enrolled in a randomized trial to reduce alcohol use. At baseline, past 90-day partner physical and sexual violence were examined descriptively and in gender-stratified regression models. We hypothesized higher reported violence against women than men, and positive violence association with HIV stigma and alcohol use across gender. Women reported significantly more current sexual (26.3 vs. 5.7%) and physical (38.9 vs. 24.8%) victimization than men. Rates were generally higher than Kenyan lifetime national averages. In both regression models, HIV stigma and alcohol-related sexual expectations were significantly associated with violence while alcohol use was not. For women, higher violence risk was also conferred by childhood violence, past-year transactional sex, and younger age. HIV-infected Kenyan drinkers, particularly women, endorse high current violence due to multiple risk factors. Findings have implications for HIV interventions. Longitudinal research is needed to understand development of risk.

Screening by Clinical Breast Examination in Western Kenya: Who Comes?

Purpose More than 80% of women with breast cancer in Kenya present to medical care with established late-stage disease. We sought to understand why women might not participate in breast cancer screening when it is offered by comparing the views of a cohort of those who attended a screening special event with those of community controls who did not attend. Methods All residents living close to three health centers in western Kenya were invited to participate in screening. Participants (attendees) underwent clinical breast examination by trained physician oncologists. In addition, women who consented were interviewed by using a modified Breast Cancer Awareness Module questionnaire. Nonattendees were interviewed in their homes the following day. Results A total of 1,511 attendees (1,238 women and 273 men) and 467 nonattendee women participated in the study. Compared with nonattendees, the women attendees were older, more often employed, knew that breast cancer presented as a lump, and were more likely to have previously felt a lump in a breast. In addition, they were more likely to report previously participating in screening activities, were more likely to have performed breast self-examination, and were less concerned about wasting a doctor’s time. Almost all those surveyed (attendees and nonattendees) expressed interest in future breast cancer screening opportunities. Conclusion The women who volunteer for breast cancer screening in western Kenya are more aware of breast cancer than those who do not volunteer. Screening recruitment should seek to close these knowledge gaps to increase participation.

Effect of a voice recognition system on pediatric outpatient medication errors at a tertiary healthcare facility in Kenya

Background: Medication-related errors account for one out of every 131 outpatient deaths, and one out of 854 inpatient deaths. The risk is threefold greater in the pediatric population. In sub-Saharan Africa, research on medication-related errors has been obscured by other health priorities and poor recognition of harm attributable to such errors. Our primary objective was to assess the effect of introduction of a voice recognition system (VRS) on the prevalence of medication errors. The secondary objective was to describe characteristics of observed medication errors and determine acceptability of VRS by clinical service providers. Methods: This was a before–after intervention study carried out in a Pediatric Accident and Emergency Department of a private not-for-profit tertiary referral hospital in Kenya. Results: A total of 1196 handwritten prescription records were examined in the pre-VRS phase and 501 in the VRS phase. In the pre-VRS phase, 74.3% of the prescriptions (889 of 1196) had identifiable errors compared with 65.7% in the VRS phase (329 of 501). More than half (58%) of participating clinical service providers expressed preference for VRS prescriptions compared with handwritten prescriptions. Conclusions: VRS reduces medication prescription errors with the greatest effect noted in reduction of incorrect medication dosages. More studies are needed to explore whether more training, user experience and software enhancement would minimize medication errors further. VRS technology is acceptable to physicians and pharmacists at a tertiary care hospital in Kenya.

Retention in care among older adults living with HIV in western Kenya: A retrospective observational cohort study

Background Retention, defined as continuous engagement in care, is an important indicator for quality of healthcare services. To achieve UNAIDS 90-90-90 targets, emphasis on retention as a predictor of viral suppression in patients initiated on ART is vital. Using routinely collected clinical data, the authors sought to determine the effect of age on retention post ART initiation. Methods De-identified electronic data for 32965 HIV-infected persons aged ≥15 years at enrolment into the Academic Model Providing Access to Healthcare program between January 2008 and December 2014 were analyzed. Follow-up time was defined from the date of ART initiation until either loss to follow-up or death or close of the database (September 2016) was observed. Proportions were compared using Pearson’s Chi-square test and medians using Mann-Whitney U test. Logistic regression model was used to assess differences in ART initiation between groups, adjusting for baseline characteristics. Cox proportional hazards model adjusting for baseline characteristics and antiretroviral therapy (ART) status was used to compute hazard ratios. Kaplan-Meier survival function was used to compare retention on ART at 12, 24, and 36 months post ART initiation. Results Of the total sample, 3924 (12.0%) were aged ≥50 years at enrolment. The median (IQR) age of young adults and older adults were 32.5 (26.6, 36.9) and 54.9 (51.7, 59.9) respectively. ART initiation rates were 70.5% among older adults and 68.2% among younger adults. Retention rates in care at 12, 24 and 36 months post ART initiation were 73.9% (95% CL: 72.2, 75.5), 62.9% (95% CL: 61.0, 64.7) and 55.4% (95% CL: 53.5, 57.3) among older adults compared to 69.8% (95% CL: 69.1, 70.4), 58.1% (95% CL: 57.4, 58.8) and 49.3% (95% CL: 48.6, 50.0) among younger adults (p <0.001). A higher proportion of older adults were retained in HIV care post ART initiation compared to younger adults, Adjusted Hazard Ratio (AHR): 0.83 (95% CI: 0.78, 0.87) though they were more likely to die, AHR: 1.35 (95% CI: 1.19, 1.52). Conclusion A higher proportion of older adults are initiated on ART and have better retention in care at 12, 24 and 36 months post ART initiation than younger adults. However, older adults have a higher all-cause mortality rate, perhaps partially driven by late presentation to care. Enhanced outreach and care to this group is imperative to improve their outcomes.

Randomized Phase IIA Trial of Gemcitabine Compared With Bleomycin Plus Vincristine for Treatment of Kaposi’s Sarcoma in Patients on Combination Antiretroviral Therapy in Western Kenya

Purpose Kaposi’s sarcoma (KS) is a spindle cell tumor resulting from growth dysregulation in the setting of infection with human herpes virus-8 (also called KS herpes virus). Advanced KS is characterized by poor responses to antiretroviral therapy and some of the chemotherapy readily accessible to patients in low-resource areas. Gemcitabine induced partial and complete regression of AIDS-associated KS (AIDS-KS) in 11 of 24 patients in a pilot study. The current study compares the antimetabolite gemcitabine with the standard care bleomycin and vincristine (BV) in the treatment of chemotherapy-naïve patients with AIDS-KS in a resource-limited setting. Patients and Methods Patients with persistent or progressive KS despite treatment with combined antiretroviral therapy were randomly assigned to receive gemcitabine 1,000 mg/m2 or bleomycin 15 IU/ m2 and vincristine 1.4 mg/m2 given twice weekly. The main end point was objective response by bidirectional measurement, adverse events, and quality of life after three cycles of chemotherapy. Results Of 70 participants enrolled, 36 received gemcitabine and 34 received BV. Complete response was achieved in 12 patients (33.3%) in the gemcitabine arm and six (17.6%) in the BV arm (P = .175). The partial response rate was 52.8% (n = 19) in the gemcitabine arm and 58.8% (n = 20) in the BV arm. Both study arms reported similar neurologic and hematologic adverse events; there was statistically significant baseline to post-treatment improvement in health-related quality-of-life scores. Conclusion The results of this randomized, phase IIA trial demonstrate gemcitabine activity in chemotherapy-naïve patients with AIDS-KS, on the basis of response rates, adverse events, and health-related quality-of-life scores.

Associations with Unprotected Sexual Behavior Among HIV-Infected Drinkers in Western Kenya

Approximately 71% of HIV-infected individuals live in sub-Saharan Africa. Alcohol use increases unprotected sex, which can lead to HIV transmission. Little research examines risky sex among HIV-infected individuals in East Africa who are not sex workers. The study purpose was to examine associations with unprotected sex in a high-risk sample of 507 HIV-infected sexually active drinkers in western Kenya. They were enrolled in a trial to reduce alcohol use. Past-month baseline alcohol use and sexual behavior were assessed using the Timeline Followback. A zero-inflated negative binomial model examined associations with occurrence and frequency of unprotected sex. Results showed heavy drinking days were significantly associated with unprotected sex occurrence across gender, and with unprotected sex frequency among women. Among women, transactional sex, alcohol-related sexual expectations, condom use self-efficacy, drinking-and-protected-sex days and age were associated with unprotected sex occurrence while alcohol-related sexual expectations, depressive symptoms and condom use self-efficacy were associated with unprotected sex frequency. Among men, alcohol-related sexual expectations, condom use self-efficacy, and age were associated with unprotected sex occurrence, while drinking-and-protected-sex days were associated with unprotected sex occurrence and frequency. Findings suggest robust relationships between heavy drinking and unprotected sex. Further research is needed elucidating the temporal relationships between drinking and unprotected sex in this population.